# Tailscale Strategy: Identity-First Networking and AI Gateways

**Podcast:** The Changelog: Software Development, Open Source
**Published:** 2026-03-11

## Transcript

What's up, friends?
I'm off the grid this week on vacation with my family.
Spring break is here.
I'm enjoying my life.
And this week I have a show for you with the chief strategy officer from Tailscale.
His name is David Carney.
We're talking about where Tail Scale is heading.
T S IDP, TS Net, acronyms all over the place, multiple tail nets, aperture, their AI gateway, click list auth, and so much more.
Big thank you to our friends and partners over at Fly for getting our back.
They support the show, they make it happen.
I'm so thankful for that.
Check them out, Fly.io.
That is the home of Changelob.com.
If you didn't know, learn more at Fly.io.
Okay, let's do this.
Well, friends, I'm here with my good friend Chris Kelly over at Augment Code.
Chris, I'm a fan.
I use Augie on the daily.
It's one of my daily drivers.
Now I use Cloud Code, I use Augment Augie.
And I also use AMP code and others, but Augie, I keep going back to it.
And here's where I'm at.
I feel like not enough of our audience knows about Augment Code, not enough about Augie, the CLI.
It's amazing.
I love it.
What can you share?
Yeah, we often say Augment is the best coding assistant you've never heard of.
And that's both frustrating as someone that works there and it's like very proud of the work we've done, but also like inspiring.
Like we want to go and and sort of punch above our weight because, like, we aren't anthropic and we aren't open AI.
And so the quality of the product itself, you know, with our context engine, once you do touch it, people are like just blown away by that.
And so, like, that keeps you going every day.
So, not to bear the lead here, but this is a paid spot.
You are sponsoring this show to get this awareness.
Now, at the same time, we're selective, and I love to use your tool.
But there is in the world, so a lot of developers look at the space and they say, Okay, well, how long can this work?
How long is this sustainable in the case of cursor or windsurf, or you pick the name and you think discounted tokens, help me shape a lens for our audience?
I think it's a lot of awareness, right?
Like, um, cursor got a lot of um publicity early on for like fast revenue growth, which well deserved.
I think, you know, frankly, some of the media got the gets the story wrong, and that like if I gave you a dollar fifty for every dollar you sent me, I'd be the fastest growing startup in the in the valley.
And so when you're selling discounted tokens, yes, of course, you're gonna grow very fast, but all that money plus more goes to the model providers.
So I think the real story is the story of anthropic and you know, being an API provider.
I think the market has just moved so fast, and there's so many pieces of competition out there that it's just hard to get noticed.
So, friends, I love augment code and I love using Augie, and I highly recommend you use it.
I love using Augie, I can hand Augie a well-defined specification, a well-defined pep, as I call them in my world, an agent flow, and it executes flawlessly.
So, the cool thing about August that I love most really is that context engine, and I can hand it a task, and it can just churn away on my well-defined plan and just never bother me and accomplish the mission.
It is so cool leveraging the latest models, the context engine, and all the fun things behind the scenes in that awesome CLI.
So, yes, go try it out, augmentcode.com.
Right in the top there is a CLI icon, a terminal icon.
Click that, install it, and change your world.
It's gonna be awesome.
Augmentcode.com.
Well, friends, we're here with David Carney, co-founder, chief strategy officer of Tail Scale.
Friends, you know I'm a big fan of Tailscale.
So David, welcome to the show.
Yeah, thank you.
I'm glad to be here.
That's a big role.
I mean, that would uh that would shake me in my boots if I was chief strategy officer of Tail Scale.
What a what a big platform you're building and a lot of moving parts and a lot of direction you can go.
It is a big title.
Uh, and I just want to be clear, I'm not the I'm certainly not the only one thinking about strategy at Tail Scale.
There's a lot of moving parts.
But being the top of it all, what does it take to what does it take to lead strategy inside of an organization like Tail Scale?
Uh well, I wouldn't even say I'm I'm leading the like the holistic strategy.
Uh, the stuff I'm working on in close partnership with with my co-founder Avery uh and a new VP product and other parts of the team.
Um, I'm focusing largely on, I guess, the strategy at the edge of Tail Scale, which is something that's sort of come about in the past year or so.
So for for the uninitiated, then, uh describe tail scale from the non-edge and then take me to the edge on what that means.
The simplest way to think about tail scale, and sometimes people ask me, like, well, what are you building?
What do you do?
Uh and so explaining it to like a lay person is very helpful.
And so first and foremost, Tail Scale makes it possible to connect any two devices anywhere in the world with strong guarantees of the identity of the user and the device at either end.
Uh and if you can do that for any two devices, you can do it for an arbitrary number.
So you can start adding like one device, one user, one server, whatever at a time until you have basically what looks like a mesh network and it's completely private.
Uh then you can layer on things like access controls.
So you can be like, oh, only these people should be able to access like these servers or devices.
For instance, like the engineers can only access production, the accountants can only access the finance servers, so on and so forth.
Uh, but fundamentally, what you can do with tail scale is create private networks.
And so when we launched the product, um, I guess we launched it maybe six and a half years ago now.
Uh, it was pitched as a VPN alternative or zero trust kind of replacement.
It does a lot more than what a VPN is, but at the heart of it, it's a connectivity platform that lets you build private networks that are fully secured and sort of using a mesh overlay pattern.
So that's the core of what Tail Scale is, and it has been for a very long time, and we've been continuing to build and build and build on top of that.
There's this motion that's uh started in the past year or so where the uh I guess the use cases for Tail Scale have gone from just internal within a company.
So it's like I'm using it to replace my VPN, or I'm using it to access production, or I'm using it to deploy like infrastructure within my company.
So that people are starting to use TailScale to deploy infrastructure to their customers.
So for instance, there's a couple of cloud provider startups that are using us in a way that they bring up a tail net is what we call it per customer.
And so they connect a bunch of like, say, bare metal GPUs with customer infrastructure and they spin up one of these per customer and then they manage it.
Um there's this ability to create multiple tailnets inside of your organization now.
So people are starting to do that to build, say, like a staging tail net or a testing tail net or a production tail net, that kind of stuff.
What I've been working more on over the past, I guess, year now, is building, I guess, applications and services on top of this platform.
Right.
So showing people like, oh, you can create a tail net.
It has these very interesting primitives where identity and connectivity and security are baked in.
Well, look how easy it is to build like applications on top of these tailnets and deploy them within a bit of private infrastructure.
Uh and that has gotten me, I guess, more involved in things like a gentic workloads and that kind of stuff, where you want really tight boundaries on who can do what with identity associated with all those actions and some kind of compliance trail and all that kind of fun stuff.
This uh this fringe, this edge, you call it the edge, I called it the fringe.
Um what are some of the things that you've thought of?
Like, how did you go with either yourself or other folks on the team to sort of go into a room and think, okay, what is the true edge?
What are the applications we can build on top of our own platform?
I'm assuming that's how you've how you position this.
Um, what are some of those things?
I know we mentioned, you know, authentication is one of them, obviously.
So you have your aperture product you recently launched, but what are some of the uh edges of that?
What products are you thinking about?
Well, the thing we started with earlier last year is that we revived a project, an internal project called TSIDP, which, like a lot of things that we've built at Tailscale, we don't do a great job of of telling the world about.
TSIDP was in our community projects repo.
Uh we'd done a bunch of work on it a year or two prior.
And for those of you who don't know what it is, and it's probably most of your listeners, that's me too.
Yeah, yeah.
So uh so TSIVP, you can you can think of it as a reflection of your identity provider inside of your tailnet, right?
So you can it's almost like a locally hosted version of your identity provider that's private to your network.
The way it works is that it leverages the fact that every connection in Tail Scale has your identity baked into it already.
When you provision a tail net, you basically have to say, like, oh, I'm gonna authenticate with Azure or G Suite or Okta or whatever.
We don't have our own IDP, we just hook into all the ones that are commonly used out there.
Well, once we can start generating keys based on like a handshake or an interaction with your external IDP, every connection has got your identity baked into it.
And so if you're sitting inside of a tail net, you know everything that is connecting to you.
And so you can actually build a small little application that says that just knows everything or know it knows the identity of everybody.
And so with that, you can actually create effectively an OIDC provider.
So that's what TSIDP is.
You can think of it as like a locally hosted private OIDC or OAuth endpoint.
And that allows you to do all sorts of neat little things.
Like you can start plugging MCP clients and servers into it.
You can build little gateway patterns where if you need to do like token um like token exchange, or if you need to do dynamic host like dynamic client registration, you can basically do it with TSIDP.
So you can keep all this interesting identity management stuff private to your tailnet, not expose it to an external IDP.
That's interesting.
I mean, uh we've talked about uh OIDC recently uh with Nicholas Zekis around NPM's security.
And that's one of the things that MPM required uh, you know, modern maintainers of this age to essentially have one blessed way to publish to NPM.
And they had this issue with like rotating keys and that secure layer was largely brought on by OIDC.
That was the first time I'd started to dip my toe into.
I'm not on authentication uh nerd too deeply.
Besides, I like to you know authenticate with things.
I like to have an idea identity when I go around.
I like my SSH keys, I like to be me where I need to be my me.
But I've I think I've been like hitting rocks together comparative to maybe what, even though I've been a tail scale user for so long, I feel like every day I learn something new about tail scale.
So, what is this?
Help me understand what that enables then.
What kind of applications does it mean?
So if if when I authenticate and I have a tail net that gives me a mesh network across whatever device I'll want to connect to, whether it's a home lab or an enterprise or a prod or stage, like you mentioned, what are some ways that enables developer to not have to like shell out to somebody else's OIDC, but to be their own within their own tail net?
What does that do?
Yeah.
So uh well, so for instance, I have a home lab.
Uh I have a Proxmox server on it.
Fantastic.
And so when I when I first started using Proxmox, like, yeah, you know, I would I set it up on Tail Scale.
I'd hit the internal IP on it so I could access it over my tail net wherever I was in the world, but I'd still have to log in with my credentials.
Um, there is a way to set up authentication with Proxmox that uses TSIDP so that uh it basically just automatically authenticates you just by virtue of the fact that you're on the tail net.
It knows who you are, right?
So I don't have to, you know, type my username and password into a modal and hit submit.
I basically just visit like my Proxmox instance locally.
And I'm in the I'm listening very closely now.
Yeah.
So uh Alex um yeah, published a lot of videos on all tail scale tech.
He actually has a video on this, and I think an update for it too, after we brought T SIDP up to the MCP spec last year.
He refreshed his video on you know how to set up TSIDP to work with Proxmox.
Uh but yeah, it's super convenient like that.
At Tailscale, we use it internally.
Um for instance, we have it set up so that uh our revenue team, if they need to access Salesforce, they don't have to type a username and password.
They just they can visit, uh they can just visit Tailscale, like Salesforce.
So this is like clickless login essentially.
Like this is just I'm me, and I can just go around my do my business inside my business or in my home lab, which is kind of like a mini playground for most people, and not have to uh even rely upon something like one password to put a password in.
You're just you and you just go there and you're just logged in.
How does that work?
Yeah, no, it I mean you're already logged in.
Like I guess that's the thing with Tailscale.
Like when you're on the tail net, you've already done an OAuth flow.
Right.
And so why do it again is basically like the question we asked.
So like every every connection can assert your identity.
So we just leverage that.
So you obviously there's a bit of work you have to do, say with Proxmox or whatever to set it up and point it at TSIDP.
You've got to set up TSIDP inside of your network.
Uh it's pretty straightforward.
Uh, we're working on making that easier too.
But once it's set up, it's just like this magical experience where people just forget about the fact that they need to log in anymore.
Uh but under the covers or under the hood, it's all you know, it's all safe and secure.
It's just like you're doing an OAuth flow, it's just silent.
So if I had a Proxmox server or maybe a database server or um, you know, an Incus uh on top of like a Proxmox V just anything like that where I have a network of other machines.
What is the process to support TSIDP?
Like is that well published?
Is that burgeoning?
How mature is that for a developer to pick that up today and just start implementing that in their in their infrastructure?
Yeah.
Well, it is uh like the code is open source.
It's on our community projects page.
Um I think it's github.com slash tailscale slash ts IDP now.
Uh if uh it does support, I believe OAuth 2.1 nowadays.
We brought it up to speed uh basically to keep pace with the evolving MCP spec last year.
So we added a bunch of other stuff to it.
Uh but people that are using it, you know, a lot of home labers are already using it.
Uh we've got a bunch of what we call TScale insiders that are that are using it too.
Yeah.
You know, again, we we just haven't talked about it enough, so I don't think enough people are using it or even aware of it.
But it's it's there and it's open for contributions too if people want to add or extend or make more requests.
Is this something that uh is it just inherent of you using Tailscale that just comes along with using TLScale and authenticating to it?
Or is this you know the OIDC part of it does it have to be like a hosted server locally and run that you authenticate against?
Yeah, you you you you need to run it as an instance.
Uh I I have I have it running in LXC on my Proxmox server, for instance.
That just like starts on boot.
What's involved in that?
Like is that running on Ubuntu?
Pretty pretty common, pretty easy via systemd.
Can you walk me into some of the details potentially there?
I think uh I think I'm using Alpine for it.
I'd have to double check.
Um but it's it's a very it's just a Go binary, effectively.
And so yeah, what uh TSIDP is built on top of this other little piece of tech we have called TSNet.
Um you can think of TSNet as a complete user space stack, like tail tail scale stack.
It's like a user space networking stack.
It's again, it's a Go library.
So you can compile TSNet into existing Go applications, for instance.
And we've done a bit of work on bindings for other languages, but the the Go chain is the most mature by far for this.
Yeah.
But what using TSNet lets you do is any Go application where you've compiled it in, you can have it show up that so it looks like just like a node on your network, just like any like laptop or iPhone or whatever.
And so it shows up as a node, gets its own IP address in the CGNet range, which we use internally.
You can apply ACLs to it, you know, policies, all that kind of stuff.
Um you can build applications with that.
That's what Aperture is as well.
It's fundamentally a TSNet application, right?
So it just shows up as a service, right?
And so the nice thing about TSNet is that, yeah, you can turn any kind of service into effectively what appears to be a device, right?
And then you can like you can apply rules in terms of who can access that with what level of permissions and all that kind of stuff.
I feel like you're the most coolest underknown crazy tech.
Like every time I peek behind the scenes even further, you know, I've talked to Alex many times, he and I are friends, uh, Alex Kreshmar, our audience knows Alex, and obviously you do too.
But every time I go a little further and a little deeper into my uh networking nerd world, I just learn more about what I can do with Tail Scale.
And it's so wild how like I'm really like when we're done with this, the first thing I'm gonna do, it's uh it's a Wednesday.
I'm hoping maybe by next week, I will get this spun up and I will not have to authenticate to my Proxmox.
What does a what is a tool like Proxmox or other services that one might run either in their enterprise infrastructure or in their home lab, which is kind of a snapshot version or a playground for most folks that might be not so much enterprise, but team based or just running ops and infrastructure?
Like who isn't doing that these days?
A lot of people are.
What does it take for a Proxmox to support that?
Is that something that Proxmox has to do?
And you know, what are some of the protocols required to uh to support this?
In terms of just running like like TS IDP as a service inside of your network?
Well, like how does how does Proxmox be able to support TSIDP?
Oh, it's just it's just another OATC endpoint.
Right.
So there it's there's built-in wall.
Supports OIDC, then they're good.
That's the that's the ticket in.
Yeah.
Or yeah, and we've added uh, I mean, there's always OAuth 2.1 support or 2.0.
I mean, we brought it up to the almost OAuth 2.1 as the MCP spec was evolving last year.
And then we pause it a little bit.
But yeah, OADC, OAuth 2, it should just work.
That's it, huh?
That seems pretty simple then.
I've uh I've never messed with this at all.
I'm gonna miss this.
It's a new world for me on that front there.
Yeah, please.
So this is the feedback.
Yeah, I'll give you feedback.
Um good.
So this is an example of the edge.
So you got TSIDP.
Uh as long as you support OIDC or OAuth 2 or 2.1, then you can use this as an OIDC provider, which essentially is the effectiveness of sign in with Google or sign in with GitHub.
Is that right?
Is that what I'm is that what you're putting down?
Okay.
Yeah.
It's like well, one of the reasons we um we spent some time working on TSIDP last year is because a lot of the existing IDPs, like the big ones out there, didn't support some of the things that MCP was calling for, like dynamic client registration, for instance.
Um, so what so we built that into TSIDP, so it effectively let us bolt on these missing capabilities, right?
So you can you can not only continue to use your existing external identity provider, but you can augment its capabilities with TSIDP inside of your private network.
Take me into that world, because there's a lot of folks who are down with MCP and uh not down with MCP.
They say it's a fad, they say it's here to stay.
I think it's all about how the context window gets impacted to the user.
I think everything in this world is burgeoning and like what was to yesterday is not tomorrow.
What is, I mean, you've been steeped in this for probably 12 months or more, building aperture and this edge you're talking about.
Uh what's your stance on MCP from uh from an implementation standpoint and leveraging it?
I was getting deeper and deeper into MCP last summer, and definitely into the fall, going to a lot more conferences, talking with many more people, and it seemed like things were just getting bigger and bigger and crazier and crazier all the time.
Uh it seemed like iterations of the spec were coming out, uh, people were reinventing things, or and it got to the point where I think a lot of companies or organizations were like, you know, we're we're just not going to implement this right now because we're worried the spec is going to change again.
It's too fraught, yeah.
Yeah, and it got, yeah, it got very fraught.
Uh and we actually pulled back a little bit from it.
Uh, because I think in September I could definitely feel this fatigue just creeping in.
Um, I was talking with more people and they sort of started pulling back from going to conferences related to it.
I think everybody was just getting tired of trying to keep up with the spec in the evolving landscape.
Um, I think uh Simon Wilson had this great quote, uh, which is a lot of people were just adopting MCP for lack of uh their own AI roadmap.
And so I think something along those lines.
I forget the exact word, but I was paraphrasing it.
But I got that sense too.
A lot of people were sort of chasing this thing because they think they thought it was the right thing to chase for them to actually come up with an AI strategy.
And it just got more and more complex for them.
And I think a lot of people started pulling, like you know, pulling the shoots, pulling the cords, and slowing down to regroup a little bit into into the fall.
And that's sort of how we felt too.
It's just like the more we sort of dug into it, the crazier it things seemed to get.
So we're like, you know what, we just need to take a step back and simplify our own thinking and focus on like a couple of problems that we want to solve as opposed to trying to chase all of them, which is what MCP felt like it was doing.
Now, in my experience with these kinds of standards, there's usually like this explosion and then things coalesce to something a lot more sane.
I think that is going to happen with MCP.
Um it's just gonna take us more time than we thought.
Dig me further into the the dynamic term that you mentioned that I get to write down with MCP.
You said it was calling for this this feature set that wasn't there, and that's why you had to go that route.
I mean, dynamic client registration, DCR.
Yeah, and I will admit I I don't know all the technical, yeah.
I don't know all the technical details of this.
Okay, so you'll have to forgive me on it.
Well, don't go to the details necessarily about that, but like what does it do?
What does it enable for the MCP certain?
Like, why is it in the protocol?
It basically lets things like MCP clients and servers uh wake up and register themselves against an uh like an endpoint without requiring a lot of manual steps or human interaction.
Okay.
So it it removes a lot of friction in terms of getting things like MCP deployed across an organization.
And you need to support that.
And that's why you went that route.
Yeah, well, MCP was calling for it.
Right.
And so it was like, well, this is this is where the spec is going.
Uh and there's there's other implement, there's other parts of like OAuth 2.1, uh, and even some experimental stuff and a few RFCs that people were referring to.
Dynamic client registration was definitely one that a lot of people were talking about and were like, look, we can build this.
Like it's not exactly rocket science.
Um it would be hard for an existing identity provider to retrofit this in because being an identity provider is a horrendous amount of work.
We don't we what Tailscale doesn't want to be an IDP.
Uh we we've known that from the from day one.
Um, but we can extend the functionality of existing ones uh in DCR was pretty straightforward for us, largely because uh with Tailscale, like you know the identity of things on the network.
So, you know, if you have this, if you have this trust in these assertions you can make in terms of like who is connecting to what, then client registration becomes a lot more straightforward that way.
Is it the point of um dynamic client registration to enable the MCP server to spin up whenever you launch it or initiate it or instantiate it to attach itself to the TLN?
Is that the point of this dynamic client registration?
Or is it the individuals coming into the MCP, maybe through a CLI or other agenc uh workflows?
I mean, in my limited experience with this again, it's been more about us just allowing.
You know what?
I don't have a great answer for you on that one.
That's okay.
That's okay.
I'm worried I'm gonna.
I'm trying to pick it up with you because this is, you know, when you're in the land of burgeoning, right?
You you kind of have to navigate some seaweed and uh some tall grass and you gotta get your hatchet on.
You're like, you know what?
I don't really know where I'm at right now.
This is moving so fast.
What is the point of this uh this need?
That's what I'm curious about, because it seems like that might be the case.
Like I see a lot of folks delivering a CLI and a MCP in one.
You know, a lot of Go applications are doing this or Go CLIs are doing this, where there's they'll they'll deliver a CLI and an MCP server in one single thing, and you can launch it via the CLI, which is pretty easy.
And I'd imagine you want that thing to authenticate with a tail net.
Like if I'm spinning up a network, I want it to, I want it to have an identity.
Be the MCP server for XYZ server or service, and that's who you are, and you've got ACLs and identity attached to you.
Yeah, and I mean that's that's how we've been showing it off to people.
It's like, oh, I'm gonna I'm gonna create a server, I'm gonna create a client, I'm just gonna have it automatically join the network.
I'm gonna, you know, a big part of again, tail scales, like you want to give these things identity, you want to pass them around.
You don't want to have people to have to say make a manual or static configuration with this kind of thing.
Like I should be able to spin up, I you know, if I'm on a tailnet, for instance, I should be able to set up uh an MCP server.
I should be able to launch it.
I should be able to grant access to that to certain people.
I should be able to tell like my colleagues and maybe some agents that are in like a tail net somewhere, like, hey, you've got this new resource available, you can go access it safely and securely.
Um that's why dynamic client registration is like super important for for us in general.
Um, and I think it's why it's so important just to the MCP spec as well.
Take me back into TSIDP.
You mentioned not wanting to be an IDP.
These are acronyms, everybody, okay?
I I think it's uh what does IDP stand for?
Identity provider.
Something.
What does that mean?
Oh, identity provider.
It's it, I mean Identity provider.
Yeah, like G Suite, Azure, Okta.
There's a lot of them, like key cloak.
Um, a thing that that, you know, uh third-party service that you trust to assert the identity of of other things, right?
Right.
Or yourself to other services, broadly speaking.
So, you know, everybody logs into everybody, well, not everybody has a Gmail account, but everybody logs into services, like they enter username and password, or you know, configure like login with SSO or you know, lock in with Facebook or whatever.
It's these are all authentication flows.
Um, something has to know uh who you are and be able to assert that to other things that people trust.
That's basically the job of an IDP.
Here's where I'm going with this.
And maybe this is just hypothesizing this edge that you're maybe navigating with you and your team that are uh thinking through these things.
I'm thinking like if you don't want to be an IDP, that makes sense, but you want to enable folks to create their own IDP, which is TSIDP.
It's this thing I can use to build my own, essentially, and I can sort of carry my own identity around with me.
It seems like I could run my own instance in my home lab on my Proxmo server, but it could be me everywhere if I wanted it to be.
That's what it seems like I could be.
Or I can build my own thing on top of that.
I feel like we're going into this world where maybe the next layer of a lot of folks' stacks, whether it's internally in a home lab, whether it's uh a small team but in the next big thing to maybe a medium-sized team that's already got motion in place, and now agentic is thrown into this world and they're bolting on and kind of rebuilding their platforms on top of essentially AI.
I feel like this world is moving towards self-hosted.
And the reason why I'm asking this question is like, is the idea for TSIDP to enable me to self-host my own identity provider?
So I don't have to log in with Google or log in with GitHub or log in with whomever, because the thing I forget most is like which one did I log in with, and now I've given you my stuff, and I gotta trust you with my SSO, my single sign-on provider, all these acronyms in this world is just ludicrous, basically.
But is that is that there am I picking up what you're putting down?
Is that the direction you're trying to take things?
Kind of.
Uh so so just to be clear, like we're like you still need an external identity provider when you're using Tail Scale, and specifically if you use TSIDP.
Like it it leverages the fact that you've got an external thing that you trust.
Because that's that's what generates the identity, and and that's what we use for the encryption key.
So you can figure out like, oh, this thing is connecting to me.
I know who it is.
Um however you're using Tail Scale right now, you've got to, you know, when you create that tail, yeah.
Whenever you're when you create that tail net, you're plugging it into whatever identity provider um is out there that you currently use.
Uh what I think is really magic about TSIDP is that it lets you not only manage identities sort of privately and internally, so you can bring RDC to all of your internal apps, you don't need to configure them to go external.
Uh it lets you start thinking about your network um as more of an extension of your identity, not just individual devices.
Like, and so you can actually start treating like this, like a tail net as uh you know a collection of identities, or perhaps just one identity.
And so it lets you, yeah, I guess have a pocket of identity that's privately your own that you can start to do and manipulate and share things in the world with.
Now there's a lot of I think there's a lot of interesting ideas and maybe a philosophy we can get into on this kind of stuff.
And I'd love to explore it.
And I know my co-founder, Avery's thought a lot about this kind of stuff too.
It's very early.
Yeah it is for us.
Right.
It's a very interesting and academic piece of tech.
Like in this is, I guess I can talk more about some of the journey last year if you want, but we started going to conferences and talking about TSIDP to people, uh, especially in the AI space, because we were like, oh, this is very fascinating from an MCP point of view.
That's you like facilitates ease of use within a side of a tail net, it helps you keep things private.
Identity is a first-class thing.
TSIDP is a way of showing that off.
And people were very interested in it.
But then they kept on coming back to just more concrete problems of like, oh, I need to get access to a customer network, or I'm dealing with API keys, or like much more tangible, like you know, first order problems.
Whereas TSIDP is like three or four steps down the line for them, which is why we pivoted last year, sort of a little bit away from it.
But uh, like as in terms of projects, it's still very actively, there's a lot of active interest, and we use it internally, but the stuff I've worked on is shifted a little bit.
Big thanks to our friends at NordLair for sponsoring this episode.
So you two a fate your GitHub org, you rotate your API keys, maybe you run Dependabot on every repo or something like it.
And then you onboard a contractor by sharing a VPN config over Slack and forget to revoke it four months after the contract ends.
And that person still has a tunnel into your internal systems, maybe even right now.
Go check it.
All this from a laptop you don't control on a network you can't see.
Well, Nordlayer is a network security platform built for businesses that actually operate the way modern teams do, distributed, remote first and moving fast.
It combines VPN, access control, and threat protection into a single platform based on zero trust.
Only the right people get access to the right resources under the right conditions, no implicit trust, no access lists.
First, it deploys in minutes, not months.
Nordlayer runs on Nordlinks, their VPN protocol is built on top of WireGuard and works across every platform.
Mac OS, Windows, Linux, iOS, Android.
No hardware to rack, no complex configs.
You can get granular control over who accesses what, from where, on which device, and when that contractor's engagement ends, well, you know what?
Revoke access from day one on one dashboard.
And it's done right then and there.
So plan start at eight bucks a month, get up to 22% off Nordlayer right now, yearly plans, plus an extra 10% off with the coupon code changelog 10 Nordlayer.
Try it risk-free with a 14-day money back guarantee.
Check it out at Nordlayer.com slash the changelog.
Again, Nordlayer.com slash the changelog.
What are the things that are most active for you now then?
Like I think this might be one of them and given aperture's announcement, and that's this is sort of the underpinnings of all that.
But like, what are some of the other things that are more active?
Like even with API keys, and those are being thrown around everywhere.
And Tailscale is kind of to some degree solved most of that.
And like you'd mentioned it's hard to tell everyone about the cool stuff you have.
And now you have a chance.
Yeah, and and thank you for that.
Yeah, so aperture is uh definitely the evolution of a lot of that exploration last year.
And so for you know, for those of you who aren't aware, Aperture is basically an AI gateway uh built on top of TSNet, which I mentioned earlier that works inside of your tailnet, and you can expose it ext uh well, there are ways to expose it externally, but essentially it's a private AI gateway uh that lets you consolidate all of your API keys inside of it.
Um and just looks like a node on your tailnet like any other would.
Uh, after spending months and months going to various AI conferences and showing off TSIDP and just talking with all sorts of people, like engineers, CISOs, people in IT, um, what have you?
Time and time again, people were saying things like, oh, TSIDP, it's super fascinating, interesting that you guys are working on all this kind of stuff.
Like, I see the merits of tail scale, multi-talnet, which is this other thing we've been working on internally, is like super neat.
But what I'm really struggling with is just trying to figure out how to manage API keys because they're all over the place.
I can't claw them back because it'll potentially like disrupt production or some engineering workflow, you know, we're trying to go really fast as a business.
Um, and uh, you know, it's just dangerous.
You've got API keys all over the place.
People trade them, they get exfiltrated or they get checked in, and they have very large, I guess, well, accounts or credit cards associated with them.
And so it's very hard in some cases to track usage because a lot of API calls are inherently anonymous.
We were sitting around at a company off site back in November, just talking about all the things we've been learning and and over the past few months, and we're like, well, wait a minute.
If if we have if we built a gateway and we used TSNet for that, the gateway already knows exactly who you are because everything that connects to it over Tail Scale has identity baked in.
So if we put the API keys all inside of the gateway, then you wouldn't need to share an API key with anybody inside of your company.
You could just say, oh, if you have a coding agent, just point the coding agent to use the gateway instead.
Gateway knows who you are, right?
So then all the API accesses uh have identity associated with them.
All the API keys can be with withdrawn.
And so that you end up with a single point of, I guess, observability, control and access for your entire team.
So engineers get onboarded faster because they can just tell their peers, like, oh, just point your coding agent at the proxy, everything just works if you're on Tail Scale.
That's the security team say, like, great, we can get rid of all the security, like all the API keys that are all over the place.
We can just tell people, just go to HTTP.
For us, it's HTTP colon slash AI.
Uh, and then every API call has your identity associated with it.
So you can just log everything, right?
And so we use Aperture internally right now.
And I think we've got, I don't know how many tens of thousands of API calls across like a big part of the company at this point.
We've just got every every interaction with all of our coding agents like going through Aperture right now internally.
So we have full visibility into how people are using AI, like all the like all the requests, all the responses, like the full logs.
We're mining it for tool calls, you know, and our security team doesn't like gets to review it if they need to, and we're working with third-party providers to like start doing analysis of the logs like in real time and sort of after the fact.
Like there's like all this like stuff that is unlocked for us, but initially it's just started with like the simple idea of like we just want to solve the API key problem.
And it's if I'm sort of I'm meandering a little bit, but one of the reasons I love it so much is it builds so much on top of the fact that Tail Scale makes things so much easier because it takes identity and encryption, it brings them way down into the stack, like right at layer three.
Uh, and if you do that, it simplifies can simplify so much stuff on top of it.
And an aperture is just an example of that, right?
Anybody can take TSNet and build an application that looks like a node on your network, and you get identity and connectivity and security baked in for free, right?
And that's and it's just like it's such a joy to work with it because as a developer for years, I spent so much time and heart, like just I had so many headaches about building authentication systems and managing you know, managing infrastructure and opening up firewall ports and dealing with like like whitelisting IPs, and now with Tailscale, you just don't need to do that.
And then applications you build, yeah.
And with applications you build with TSNet, you don't need to do that.
Right.
So it's um, yeah, it's just been a joy to work on this project over the past while as a home labber.
I'm I'm thinking about uh having is aperture available to anyone, yeah.
Yeah, is it a product?
Do I have to pay for it?
How do you deliver it to someone?
It's I mean, it's an early alpha, okay, right now.
Uh there is there is a self-serve flow.
We launched it just uh well, we quietly launched it a couple of weeks ago.
Uh it's not quiet now.
Yeah, yeah, yeah.
We did we did a a little, we did more of a push just the other day on it.
Right.
You know, but if you go to aperture.tailscale.com, um you can sign up.
Uh there's a bit of like we manage it as a wait list just because we don't want the servers to get overloaded.
But the idea is that we're gonna like open it up like very quickly for everybody as soon as like we're sure that things are just gonna scale just fine.
But basically, you know, we will provision an instance, you know, you authenticate it as a node into your network.
Uh it just shows up, you can start using it right away.
Um, it's you know, in line with what we're what we already do for Tail Scale, like it's you know free for home lab use.
Well, like it's bundled, it's we're gonna be announcing how we're bundling it and as part of the free plan, just you know, free for home use, that kind of thing, just like we do.
Um obviously, you know, we're planning on it being a paid product for enterprise, you know, but we're still exploring pricing and and all that there.
But yeah, yeah.
But I want you know, I want every home lab, you know, anybody who's playing with LMs and API keys and stuff at home, they should just be using it.
It just makes things easier, it's sort of like the the tail scale way.
Is this self-hosted then?
Or is it not self-hosted?
Because you said provision and instance.
Yeah, so we are we are hosting these instances for customers right now.
There is there is plans and talk about self-hosted versions, and certainly enterprises, like some of them would insist on that.
There's varying degrees of what that might mean.
If terms like customers might be like, oh, I want to bring my own cloud, you just write the logs there, but you can host the actual, you know, the stuff that's taking up the CPU.
Or some customers might want just like, no, we have to have everything on prem.
Yeah.
But but right now, we wanted to get aperture into as many hands as possible as quickly as possible.
And the easiest way to do that, and I think one of the safest ways, frankly, is just to let us host the instances at this point.
The way the reason I asked that question is one, I said earlier that uh I feel like the world is moving to self-hosted.
As a home labber, I already felt that way years and years ago, but I feel that way more and more now because when the cost to produce an application that's bespoke to me goes to near zero, except for my uh ability to specify it and my ability to uh describe intent, then everyone theoretically can be a builder in this new future we're going towards.
And why not self-host a lot of things that I'm gonna do?
Because I'm already a home labber, that makes a lot of sense.
But I imagine that a lot, a lot of teams, a lot of tech companies, a lot of non tech companies who just like that are now tech companies, they think the same thing.
They want uh sovereignty over their things.
They want to control their CPU costs.
They want to trust the cloud less and you know, still leverage cloud native type things, but in their own controlled way.
Especially when it comes to identity, and especially when it comes to all tool calls and all responses, etc.
I mean, because as an individual AI user, a team of one basically, when it comes to like the things I'm building, uh, one of the things that I have anxiety about is, or just I suppose not anxiety, but like I just wish there was a record.
It sounds like with Apatra I can gateway my way into all my AI and have my prompts and the responses stored there versus the compacting that happens and goes away.
And you even have, you know, in Claude Code, for example, you have an export where you can export the the conversation basically.
It's like let me snapshot what we've talked about.
So worst case I could walk away with context of the conversation, maybe not context of what we actually the underneath we've described and and you know some of the world we explored.
I feel like that to me, like I would want personally, maybe this is direct feedback, but like as a home labber, I would want to self-host that especially because of how um secure or um exposed I might be, you know, with with those it's like uh not self-hosting your email.
I think today you definitely don't do that.
But I think in this case it's such sensitive or could be such sensitive information.
I personally would prefer to self-host it.
And I'm curious why, given that you largely haven't done a lot of infrastructure in the history of Tail Scale, like you've pretty much been a pointer in a lot of cases, and not a lot of infrastructure required to build what you've built.
Why now?
Why build out instances and hosting and I guess responsibility.
Yeah or or liability is liability too.
All the all the abilities, you know.
Yeah.
No, it's it's a very accountability, responsibility, liability, yeah, all those things.
Yeah, no, no, it's it's a very it's a very, very good question.
Um, and I agree with you.
I think a lot of people uh will want and will expect to self-host it.
And we do want to provide a path for that.
Yeah, it's just yeah, take for the purpose I get that.
Yeah.
It's early.
It's early, and it's you know, this little team of mine right now that build aperture over the past few months.
Um, we're up it's cliche, but we're operating very much like a startup.
And it's like, and it's like, well, we just need to choose one path and go really fast on that path.
And it's like, what's the quickest way we can get feedback from customers, get people, you know, get people experimenting with it, getting to the point where we actually have like traction.
Like, you know, it's it's do we have product market fit?
Is the fundamental question.
Once you have more product market fit, then we can sort of we can start branching out into figuring out where that takes us, right?
But this this uh, and frankly, internally, uh I also wanted to experiment like, well, you know, I think there is a world, to your point, I think data sovereignties and data governance is extremely important to people.
Yeah.
Um, but so is speed.
I mean, it's I think there is a lot to be said for motions where it's like, oh no, we can like tail scale can help you provision instances.
Like maybe we hook it to other cloud cloud providers, or maybe we get to the point where I don't know, like people could start sharing their infrastructure and to compute with other people, and we can make it possible for people to like, I don't know, create their own cloud providers or some kind of like democratic cloud system using tailfoot tail scale infrastructure.
But I wanted to experiment with the idea of like, well, how what would it take?
Or how much would customers resist a like, I just want to launch an instance into my network, like just make it easy for me.
Like just give me a one-click, drop, drop a node in, right?
Because there are still the guarantees, like we can't just simply start injecting nodes into customers' networks, like people have to authenticate them in.
So there's a lot of guarantees that we already have in terms of our ability to just or the lack thereof to I mean take over customers' networks.
Um I think a lot of customers just want the speed and convenience just to try something out.
So we made a decision.
It's working out really well so far.
I've been actually surprised with how many people have just been willing to say, like, yes, spin it up, add it to my network.
You know, at some point we may want to have a conversation about self-hosting, but for the purposes of experimenting with it now, like this is totally fine because we trust you guys.
But yeah, it's really days for us too.
You know, there's something else I'm this thread I'm pulling on to with this self-hosted world is like, you know, especially in your history, and just just go with me here and think out loud if you don't mind.
Put all your cards on the table, David.
Um I've been thinking about this world of self-host, like we just had a uh a really good conversation with the founder of TL Draw.
And one of the things that TL draw basically does is sell an SDK.
So it's not even you know, fully baked software, it's it's the SDK to build on top of.
You know, it's the concentrate, you would just add water essentially, which is one kind of fascinating.
Great business, a lot of upside, uh, no infrastructure, you know, high margins because the there's there's no servers to spin up, there's there's no uh attached to AWS or GCP or you pick your cloud and you're you're now competing with them at some point because you got software that you're putting out there in the world as licensed software, whether it's truly, you know, uh MIT license open source or it's source available open source code, not to blend the terms there, just to be clear about that.
But you got this world where you have, I think we'll have a lot more people wanting to self-host, and those people can still be customers.
You know, have have you all explored or have you even thought about yet uh if you went that route, how you can license it and still sell licenses of it, where you can provide essentially a source available version to the open source world that hey, if you're in a home lab environment like you do now, totally free, this license applies to you.
But if you're in production or you're in an enterprise or you're in this kind of business class, the source is available.
Contributions are not necessarily uh what you're trying to achieve, but you do want to have your code out there so people can see it.
But then you're also saying here's a license to you, or here's a license per node.
You want to have uh you want to have an aperture instance in your stack, you want to have multiple.
I don't think you'd I don't see why you'd need multiple.
Maybe just one is fine, but you got uh maybe a per user license or a per node license.
You tell me, I don't know how it actually permeates when into the network, but have you explored that world yet?
Are you planning to Yeah?
Oh, we're planning to.
We've definitely started talking about it.
And even some of these bigger, like we've had a lot of enterprises show up that are very interested in aperture and they've been bringing these things up with us as well.
Uh I think I think there's a world of possibilities for us there.
I'm like, I'm a big fan of open source in general.
Uh, and I'm a big fan of like self-hosting things.
I like I like to play around with it.
I like to think of it, I like to feel like I'm in control.
I like to be able to just turn it off if I want to.
You got a home lab, David.
Of course.
I mean, that's that's where we bake all of our ideas first.
We take them, we do them in the home lab, we figured it out how it works, we nerd out, we learn, we explore, and then we take it to wherever we're going.
That's kind of cool stuff.
Yeah, exactly.
And uh yeah, I think with me is like the the bigger thing that just keeps me going as a as a founder is like I want everybody in the world to be using Tail Scale because it's just a better way to do networking.
Right.
And so anything that's gonna encourage that, I want to pursue it.
And I think obviously, you know, having a self-hosted version of Aperture is one way that that will, it'll enable that.
Getting more people just to build TSNet applications is another way to do it.
Um, so everything's on the table right now.
For the purposes of um, I guess just speed and getting feedback and figuring out where we need to go next.
Yeah, yeah.
This this model that we have right now, I think is the best one because frankly, we're making updates and rolling at features so quickly that trying to manage those with self-hosted installations and like worrying about database upgrades and all the security and everything behind them.
It just it we sort of did a bit of the math.
We're like, we need to choose one path and keep it super simple and just manage things like aggressively.
So it makes sense in this case, then to cloud first, not because you're not for self-hosting, but because you need to maintain velocity.
Yep.
And the only way you can really maintain reliable velocity in this case is to have your own instances.
But in the future, if someone says, hey, you know what, I love this, I'm down for it, except I want to self-host.
Then when the product has proven itself, that makes more sense later on.
That makes a lot of sense.
Yeah, that's that's where that's where I definitely that's where I'd like to go with it for sure.
It's easy to say that it's easy for me to say, hey, just hey David, just you know, make this a self-hosted instance, right?
I mean, it's easy just to say those words, but then to actually support it is the challenge because then you have versions out there, and I suppose you do have data-based migrations, but I mean, if you know which version you're on, you can pin to a version and you do a good job with engineering and database migrations, then that can largely be collapsed to to almost no pain.
I get that it can be painful.
Uh, and then you also have containers, Docker is is certainly one of the things you can easily do that with.
They've rolled out hardened images, so I'm sure they would easily roll you into uh you know an aperture hardened image that you can just boom, it's there, you're using the latest version of it kind of thing.
So there's a lot of things you could do to simplify it from a hosted standpoint or a self-hosted standpoint.
That doesn't have to be here's my code.
It is a binary too, right?
You say it's a go binary.
You're largely a Go shop.
So I mean, Go Binaries plus system D or Go Binary Plus Container is pretty easy worlds to navigate for the most part.
So yeah, yeah, it it is designed to be portable like that.
Yeah.
Uh I mean, I'm excited about all the stuff you're mentioning.
Uh it's it's definitely stuff we've talked about internally as well.
The moment you offer, I mean, I'm gonna try it no matter what, but I would the moment I mean I'm a self-hoster.
And in fact, uh, a little side note here, I want Alex to bring back self-host it so bad because I feel like uh now is the time to bring back the podcast self-hosted.
I don't know if he's thinking about it.
I know he's got a big job to do for you all there.
Maybe he's just too busy, can't think about it, but I like the idea of it because it's just so much happening there.
Yeah.
Um, aperture is built on TSNet.
You mentioned TSNet applications.
What are some other things?
Like, can I go build on TSNet?
Like if I'm a builder, what ideas can you give this world to say, you know what?
We can't do it all.
Here's some ideas.
Do you want me to build?
Do you want people like me to build on TSNet?
Or do you want to be the only application builder?
Oh, no, no, no, no, no.
Like the bigger arc here is like I think Tailscale it can and should become a platform.
I think every platform needs a few killer apps.
Yeah.
And it needs to sort of lead the way.
It needs to demonstrate to people like, look, this is viable.
Like this is why you should spend your time on this kind of thing, because there's an ecosystem that you can plug into.
And Aperture is definitely one of those, is definitely one of those projects for us.
Um, if somebody else goes and takes an existing uh AI gateway and retrofits it on top of Tail Scale, that's a win-win as far as I'm concerned.
Just more people are using Tail Scale, customers are have a high better security, you know, uh more people are just familiar with TSNet, which is fantastic.
Uh when you start thinking about Tail Scale in terms of like, oh, I can create a private network and I can add things to it, and then I can start controlling access to it.
And it's just like it's a very abstract uh, I guess, model.
Um, there's a lot of different kinds of opportunities and problems that can help to solve.
And I and I will say there is something that we've started talking about more in the past 12 months, uh is something called multi-talnet.
We have a blog post or two about that, which is the ability for you to create multiple independent tailnets instead of your your org or your team or your home.
Right.
And there's there's API only ones right now, which is basically for machine to machine use cases.
Like there's not a strong notion of identity, like a user identity in that.
And then there's there's ones that are more have more, I guess, that are more directly tied to the user identities.
So we've got some customers that have like a staging tailnet and a testing tail net, and then like user identity is a first class thing.
Those I believe those uh, I guess those tracks of development will converge at some point.
Uh but there's already some really incredible stuff you can do with multi-tailnets, for instance.
And that's one of the reasons I think things like Aperture and TSNet are so fascinating, and that I could create a separate tail net, uh, and I could make sure that like nothing can escape from that.
And I can go off and do its thing, and I can run coding agents in like YOLO mode, right?
Uh, I can have them connect to MCP servers, and you know, I have permissions that I can fly around inside of the network and it's all nice and contained.
Um, and you can use Tailscale for that kind of thing.
And that's what I'm those are the kinds of areas where I'm really excited for people to start experimenting with and bringing up ideas of just like, oh, like if Tailscale did this for me, or if it like let me export this kind of thing, or if I could transport this kind of policy, or if it could connect to these different things, then I could achieve X.
And I want to hear more of those ideas so that we can start building more stuff and more features at the edge so that people can start using TailScale more as a platform.
Yeah, I'd love to explore the idea, not so much in this podcast, but something that's on my plate is network isolation.
Like spinning up an instance, let's just say uh like an Incas.
Are you familiar with Incas by any chance?
Did I say that earlier?
Yeah, no, sorry.
Canonical used to have LXD.
So you know that you've got LXC in uh, and I don't even use those, so I'm not super versed.
This is a world I'm exploring and learning more about.
So I'm sure my audience can be like, Adam, get yourself up to speed here.
Okay.
Uh uh, I'll attempt to follow this here.
Incus is like system level containers.
So versus Docker, where it's a protocol, it runs on system D, it spins up instances, and it is a Go binary.
It's built on Go.
The person who invented it for Canonical, Ubuntu's uh parent company, it was called LXD.
I think they had a license snafu.
I don't want to, I don't know what happened there, but something happened licensing wise that made Canonical change the direction of it and remove it away from I believe it's Linuxcontainers.org, if I can recall.
Let me just double check my notes here so I don't jack up that.
Uh yeah, Linuxcontainers.org, which if you go there, you will see Incas, you'll see Incas OS, you'll see LXC, uh, which is something you use in Proxmox, distro builder, which can do some cool stuff with like building distributions.
But Incas, let me actually go to their homepage so I don't actually jack up what they say.
Incas is a next generation system container, application container, and virtual machine manager.
So much like you you can do a VM and you want to actually have control of the kernel, or the container itself can be you're borrowing the OS level, the hosts level kernel, where a VM you want to have your own kernel, so it can do both of those uh in easy exchange.
So they're uh unlike Proxmox, they're very much differentiated and actually largely abandoned from a CLI standpoint.
They don't really have a good CLI in the Proxmox world.
So if you want to spin up a new VM or a new LXC, you've got to do a bunch of clicking inside the web UI.
Painful.
Whereas Incas is has got a really great CLI, you can script a lot of this stuff, and the difference between a container and a VM from a scripting API CLI world is the same.
Like the same kind of commands, but arbitrarily different whenever you spin them up.
Why am I telling you this?
Okay, the reason why I'm telling you this is like is I want to I want to spin up different layers of Incas or a different container of VM.
And I want them to be in an in an isolated state.
I want that particular container to know nothing about anything else around it.
As far as it's concerned, it's in the black space.
Like it's literally, there's no star around it.
There's no asteroid, there's no planet, there's no moon.
Like it's just in a sea of abyss.
And all it could do is do internet traffic out, pull down updates, send traffic back, but to its peers, there is no peer.
You could do that with typical networking, but it sounds like with a multinet or some tailnet food that I just don't have yet that I'm still learning about, I might be able to jail one of those containers or VMs in ways I just weren't, wasn't able to before.
So that's why I'm camping out as like this world of like network isolation, put that new instance, that new VM or new container into network jail, essentially.
Basically, yeah.
I mean, a lot a lot of the stuff you can do with multi-tailnets, you can you can do with uh just modifying the policy file in a single tail net.
You know, if your rules are aggressive enough, for instance.
Yeah.
Um, but there's definitely there's a lot of people that were like, no, I just want, I don't want to mess around with a like a complex policy file.
It's too risky, or I'm dealing with multiple customers and my customers demand, you know, complete guarantees on isolation between them.
Like I don't want to be managing something where I accidentally like add a star somewhere that all the customers can see each other for instance um and so yeah multi-talnet is definitely it's it's it it it facilitates that significantly so a lot of it's just peace of mind like oh no I I have this particular tailnet is used exclusively for this it can't like it it it's not like it can make a lateral it can't move laterally to a different tailnet like they're completely isolated from each other right um so it gives a lot of people just peace of mind and frankly it just it's like a divide and conquer kind of problem it's like well why have like one big complex tailnet when I can have two simple ones or I have my own that I've authenticated with because you said you have to have an OIDC provider right I've got to authenticate with tail scale with one of those providers could be Google could be GitHub could be whomever I choose.
And so that establishes my tailnet and then beneath me building on tail scale I want to be able to have a whole separate tail net that is not I guess even tied to mine but something I can talk to but it's isolated that's where I think you provide some networking that is just like dark arts essentially.
Yeah, well, yeah, with well, with the API only tailnets, you basically when you create when you get um I mean you get an OAuth client back.
Um, and then with that OAuth client, you can do things like add nodes, create auth keys, provision stuff like within that particular tail net.
So there is it is still tied with like a what you might call the primary tail net, like it is associated with that.
Uh, but you know, for all intents and purposes, it's a separate network.
Right.
Well, for me to spin up another sub tail net, I have to have a tail net, which means I have to authenticate to tail scale, right?
So I have to have a tail net to begin with to spawn subtelnets that are basically blue, you know, they're just blue oceans, they're they're by themselves, they don't have a clue whoever else is around them.
They're just but I have to create I have to have tail scale and a tail net to create subtail nets.
Yes, exactly.
Yeah, yeah.
And you could hand roll a booking system, wire up Stripe, build an invoicing workflow, stitch together email marketing, and honestly, you'd probably do a great job doing that.
But that's weeks of your life spent on infrastructure that isn't your actual thing.
Sometimes the smartest move is choosing the right tool so you can focus on the work that actually matters to you.
That's Squarespace.
It's the all-one platform that handles the business side of whatever you're building for yourself, for someone, for a friend, whomever.
And two things I think are worth noting.
First, offering services.
If you're doing consulting, workshops, freelance dev, mentoring, content.
Squarespace brings scheduling, invoicing, online payments, and eml marketing together in one single place, one dashboard.
You list your offerings, clients book and pay, and you skip the part where you are playing accounts receivable in your DMs, and it's a real business, it's a business workflow, not a pile of SAS subscriptions that are kind of duct taped together.
Second, selling content.
If you've got your course ideas, your video tutorial ideas or deep or deep technical content you've been meaning to package up.
Squarespace lets you gate it behind a paywall, one-time fee or subscription, it's your call, recurring revenue from expertise you've already built.
That's a good trade.
And the point isn't that you can't build this stuff, it's that should you?
Maybe not.
Okay, head to squarespace.com slash changelog for a free trial.
And when you're ready to launch, use our offer code changelog to save 10% off your first purchase of a website or a domain.
That's squarespace.com slash changelog.
And we use our link, of course, you're supporting the show, and we love that.
Once again, squarespace.com slash changelog.
We got there by talking about different applications you think can be built on top of TSNet.
And I think where we may have dropped off at or not given a good trail for, and I know the listeners may be potentially upset if we didn't do this yet.
Is like, can you give me a couple ideas?
Like if you can go and if you were at a if you're in the hallway of a conference right now and you were just talking to some folks, you know what?
Here's what you can go and build on on your TS net or a TSNet application.
Here's a place you can go play.
If an enterprise came to you and and they're like, you know, we're just leveraging our tail scale, our tail net in these ways.
How what are some problems they're coming to you with that you're like, we're not gonna get to that for a while?
These are things you should build.
What what problems are some of your largest customers just you know on your uh not not so much on your case about, but what's the loudest cry in the woods of this is what I want to build on top of my TSNet?
One of the bigger issues that a lot of, especially larger companies that I've seen coming up more recently is they have a traditional network, traditional VPN.
It's like one monolithic thing.
And they're trying to bring up MCP servers and they're trying to bring up MCP clients, and they've got these notions of like they basically are thinking like, oh, I've got now agents that are trying to operate inside of the network the same way humans used to, right?
Except um, well, I don't need to go on about the dangers of sort of letting uh an agent run amok on an internal corporate network.
Um it's yeah, right.
It's becoming, yeah, yeah, it's uh it's becoming uh obviously more and more an issue for like especially bigger organizations that are that have like traditionally dealt with security from like a sort of a very centralized monolithic um I guess perspective.
Right.
So I think there's definitely been this push of just like, okay, hell, how do we start like a segmenting or isolating or subdividing our network?
How do we start decentralizing it more so that we can enable these different teams and pockets of the company to work more independently with each each other or within the like within sorry, to work independently to solve like particular domain problems while still uh I guess enabling the velocity and freedom of access to like required information, right?
So I think I think tail scale is actually a way that companies can do that.
I think there's there's a lots of different approaches, but tail scale definitely makes it easy.
It's just like, okay, well, instead of like one gigantic monolithic network, like maybe you start building like one tail net per workload, for instance.
And we've seen this kind of thing with people who are playing around with Kubernetes, right?
Right, and they have like customer internal tooling that they need, like some kind of internal and monitoring system.
Let's like, oh no, when we bring up a cluster, we've got these applications that need to sit inside of it.
It has to sort of moderate or govern like certain kinds of networking access.
Like we do a bunch of like log analysis.
Maybe we do a bunch of real-time stuff.
Like, how do we how do we insert these applications that we've built into like these networks, you know, in an isolated way?
And so I think there's a lot of internal applications inside of companies that need to stay internal and private.
Tailscale is good for that kind of thing.
You can retrofit them into some inside of TSNet, and then you can encapsulate sort of entire workloads inside of TailNets themselves.
And it keeps uh it helps people reason much more about like a safety and security that way.
Where does someone get more details about multinets then?
Because this is multi-tone?
Yeah, yeah.
We've I mean, if there's a couple of multi-tail net, yes.
Let me let me be on brand with that.
I like that.
Yeah, yeah, yeah.
Multi-tail net.
Yeah, then I'm just shortened it to multi-net because everything else is something net.
TS Net, like multi-net.
I'll just try to follow your lead there, David.
No, no, it's good.
And in fact, I'm sure our product team will watch this video and maybe and maybe think more about the name.
We've been debating it internally of how to call these, like what to call these things.
Um there have been a couple of blog posts even over the past year.
So even if you were to Google for like tail scale multi-tail net, you would find one.
Um it is, I believe, in beta right now.
Uh I'd have to double check that.
Um, but it is accessible.
Yeah.
To Helm Lab users.
Yep.
Sweet.
Okay.
Uh while you're doing that lookup here, I'll I'll fill some air with this.
I think for the branding team or marketing team, whoever's listening, uh, our listeners as well.
I think the the better name actually is multi-tailnet because you already say tailnet.
It would not make sense to like shorten that to multi-net.
I messed up.
And so I'm putting my like branding hat on thinking about how I would actually frame it in my own mind if it were my product.
And I would call it a multi-tail net, because that's truly what it is.
That's what makes the most sense to me as well.
So if that's what y'all are thinking, good job.
Yeah, I just double checked.
Yeah.
So we our marketing team did a great job with this fall update we had back in late October.
And there's a blog post called One Organization, Multiple TailNets.
It talks about this um over a page or two.
Uh, and then just how things are, at least at the time of this blog post in Alpha program.
But we've been, you know, steadily working on these features on the features related to multi-talnets over that time.
I'm super interested in this.
I'd love to see where this goes because I think there's a lot of a lot of opportunity with subnets, multi-tail nets, you know, this problem that your you know your customers are bringing to you.
I I concur with that because there's things that I'm doing inside my network that essentially is on my flat VLAN.
You know, I got my traditional network that's across, you know, my 192 space.
Um, and then I got my tail net, of course.
And there's things I want to isolate that I'm just not, you know, and I guess I'm just crossing my fingers because I I I'm moving at a velocity that doesn't let me slow down enough to be secure.
And that's the job of Tail Scale, I feel like that's where, you know, that's why I use Tailscale.
That's why I keep you know investing more and more of my knowledge into what you are doing, because you guys are like just networking wizards, and you love Go.
I love Go.
Yeah.
Uh if you're not gonna be at GopherCon this year, David, you all should reconsider that.
Uh GopherCon is the best place to be.
Um yeah, I I just think this is super cool.
This idea of multi-tail nets.
And so I want to like dig into this.
Between this and uh T S O I.
What was it?
My gosh.
So many acronyms.
I mean my notes here.
Don't tell me.
We need better names.
We need better names for these things.
It's okay though.
I mean, it is a T S IDP.
That does make sense that they tail scale identity provider.
That totally makes sense.
There's just so many acronyms in this world.
It's hard to keep them all on the tip of my tongue.
So you got a lot to cover there.
So those are two things I'm taking away from this for me personally on homework.
Uh, because I don't want to log into my proxmos anymore.
I don't want to log into certain things I'm self-hosting.
If everything I self-host that has a login prompt, even my true NAS box, if I if my true NAS box can be OIDC compliant and let me use uh this TS IDP, gosh, I'm I'm on it.
Yeah.
Uh then that'd be great, you know, because I don't have to log in anymore.
That'd be awesome if I my identity could truly just follow me.
And I think the the challenge that I think for you all is it's a constant, I would say marketing battle, but I think it's like um you have so many home labbers out there, uh, so many people probably playing with your stuff.
You need more people just naturally sharing this stuff.
And I know Alex does a great job on your YouTube channel.
I mean, just tremendous job on that front there.
Yeah, I think you're gonna constantly have this problem, David.
I don't I don't envy your position at all with constantly having to update people, but there's just a sea of great stuff underneath tail scale from multi-tail net to the things we've just talked about.
Those are things I'm walking away with.
You know, I do want to circle back to, if you don't mind, this idea of a vaper.
And really just this less about the product, but more about this idea of of having an AI gateway.
Yep.
Help me understand why you feel so strongly about it.
I imagine you do, because I can't I'm I kind of having this feeling too.
Is that I'm not sure if I'd apply it in my home.
I think maybe I would, but certainly in a burgeoning team, a small team, uh, definitely in a smaller business or an enterprise where you're sort of like mid, small to medium size.
But this idea of an AI gateway, one for security, two for API keys, and just like the the concerns there.
But even just me have not having to have this anxiety of loss when it comes to transactions I'm doing with a generative AI, where I'm knee deep in a world, we're exploring it.
You know, it's largely just pros, it's largely defining specifications, largely defining intent.
It's largely even learning about things, building toy applications just to learn, not sometimes so much to build something to ship it to the world, but to learn about, you know, even authentication.
Like when I go play with multi-tailnets later on, I'm gonna go build a toy application and I'm gonna work with an AI to do that.
Awesome.
But it sounds like the gateway can help me uh have a transactional history.
Help me understand that world of an A an AI gateway and what people can do with it.
Yeah.
So I mean, we're definitely not the first person to build an AI, like an AI gateway.
You know, it's uh I mentioned this in a blog post just the other day, but you know, I was talking with another founder like months and months ago, and they're just like, we're just talking about the proliferation of gateways, and we're like, oh, it's the obvious idea.
It's like, and I agreed at the time, you know.
But then later on, I realized, oh no, there's a tail scale spin that could make things much simpler, and so we should just show the world that.
Um but aperture in particular, uh let me think.
Maybe I should talk about it, but the reasons I love working with it.
Sure.
Because I think it'll resonate even.
You've been like knee deep in 12 months of history.
So you or research, you must have if you don't love it, we got problems.
Yeah, yeah, no, I I love it because it's uh it's it's kind of like tail scale.
The first time you use it, it was just like, oh my gosh, everything just got easier.
And it uh you know, it and tails like tail scale that seeing that wow, like or or that realization of just like, wait a sec, like here's it's um it's like here's all the work I didn't have to do kind of thing.
And aperture is kind of the same thing.
It's like, oh, I just if you have a gateway, if you have a coding agent rather, that you can point at a proxy, you can use you can use aperture, right?
Um if you're using API keys.
And so I I've run it, I mean, I I I've run Aperture on my home lab, and the reason being is just like, oh no, I've got a server, I like to run everything through it.
I like to have all my logs, you know.
Like I just I just want one spot that I, you know, if I'm working on my laptop or elsewhere and I need to connect to my tailnet, I just have a single path, something where I I just know I'm just gonna route everything, you know.
So I can take my tail net anywhere.
Like, and that's one of the the nice benefits of it all.
Uh from um from a team perspective, it lets us um it lets us have full visibility into the logs of other members of our team, right?
So I can it's like, oh, how is like you know, my teammate Ben?
Like, like how does he write prompts?
Why like like why are his why are things with him like he he does things much more efficiently than me, right?
I'm curious, like, oh, I want to learn from Ben, like how is he constructing the prompts?
How are like how are you like how is he interacting with these LMs?
Um, it lets us have access to all sorts of I guess different backends.
So we use Cloud Code extensively.
You can get to Opus like through a variety of means.
You can go directly to Anthropic, you can use Bedrock.
Um, you know, so we've got both of those configured inside of Tail Scale.
And so, like, you know, there's been times where like anthropic or bedrock have had issues with being able to just quickly like switch over to the other one.
Um, it gives us obviously I guess metrics into like token usage across the team, so like input, output, cache tokens, reasoning, tokens use using what, that kind of workload.
It gives our security team just visibility because like every most people don't realize this, like every uh every API call is stateless, right?
Which basically means that entire context window is getting shipped back and forth across the wire every single time.
Like we log every single one of those, and so and then we've got we've got some tech in there so that you can consolidate those into sessions, right?
So that you can actually, I guess, like you can go through all the API calls in a given coding session and sort of get context and understand what's going on with that, um, which is really helpful for visibility.
It's like, oh, like, you know, what was Carney working on at 2 a.m., for instance, and then from I guess maybe more of a comp like a legal and a compliance standpoint, you could actually start, you know, you could start pointing your Git histories back to individual coding sessions.
Like, oh, like this code was developed in conjunction with this developer, like and here's the proof of why and who contributed it to it and how, because I know that's an issue that some legal teams have brought up.
Uh our security team, you know, we can export the logs.
Um, there's a lot of like like post hoc analysis you can do on those kinds of things.
We're working with integration partners to do like sort of real-time investigations of like tool calls, for instance, or like like like or like uh after the fact log analysis.
Um there's a bunch of fascinating stuff there.
Could you block things?
Could you like working?
Yes.
I mean, the short answer is yes.
Something like a firewall, like an AI firewall, even too.
Yep.
And that's why I think it's really interesting.
You could do so much with this gateway.
Yes, you can.
And there are uh so we mentioned them in this in a blog post just the other day, but there's some integration partners.
Oso is one, servos is another, yeah, for instance.
Yeah.
And so uh yeah, they've been great, uh like to work with over the just over the past few weeks, but yeah, we have a we have an integration with them where uh we've got some we have an API, for instance, you know, tool called hooks, like they can intercept them, they they can analyze those, and you can start to do things like where you can dynamically adjust your network um or your security policy based on effectively real-time signals.
Yeah, like activity.
Like what's being called, what's being prompted?
What's is that what you mean by that?
Like if if I'm trying to do not so much nefarious things, but let's just say dangerous things.
Yes.
I could be a new developer, new builder, uh, or just maybe have ulterior motives.
Who knows what?
But like could you pay attention essentially and like do different things and react?
Yeah, or you or you have or you have like an agent that is starting to get a little bit more fast and loose with the rules and starting to access things that maybe.
Listening to the yeah, okay.
Yeah.
And so there's there's there's ways um, you know, we can send signals out to those tools, and those tools have like their own kind of like I guess application level like network policy that they can start to uh adapt in real time based on signals they're getting from aperture because it's all centralized, so you can say, like, oh, this thing, like this agent or this person or whatever, like they're deviating a little bit off of like the regular kind of behavior.
Maybe we should start locking down or auditing more and slowing down their interactions with these kinds of resources.
Um is there latency involved in this if you're doing like tool level?
Is it literally like are you does the call go to the gateway back to the agent to allow it?
Or is it just paying attention and sniffing it?
Well, right now, uh, it's just a sniffing part.
We actually are working on uh I guess adding an approval loop, right?
Well, like RMs or like you don't do don't do RMs that are like dangerous RMs, for example.
Like, yes, do some of those because you have to remove files and add files and take things away.
So, yes, do that.
But if I can intercept a really dangerous RM or a SQL injection or a database drop or table drop, or you know, who knows what, like I would want that, that's where I would want the gateway to even as a developer uh give me the ability to go in the dangerous zone of using the agent because I want it to have free reign.
I really hate sitting there babysitting the yeses and the no's and like oh my yes, please commit and push the code.
I don't want to tell you to commit and push.
That's inherent.
That's what we do here, okay?
It's part of like getting in and out of the door.
Let's let's do I don't want to babysit you doing that.
So I almost live in the dangerous world.
I do live in a dangerous world, I would say.
Like I do dash-dangerous a lot on pick your, you know, codex, opus, whatever.
Um, but I feel like this gateway could be my security policy and maybe even my own, my own personal big brother in a way.
Yeah, it's uh yeah, like Tailscale, we're not gonna build that entire ecosystem.
It's one of the reasons we want partners so early to demonstrate the fact, like, oh no, we want to work with a lot of people.
We're not gonna like, we're not gonna build everything ourselves by any stretch of the imagination.
Uh, there's a lot of really incredible tech out there that you know we want to be able to plug into.
Uh a lot of great teams doing like deep research on this kind of stuff.
Like we we think of our tail scale and as generally like a very broad horizontal like connectivity platform.
It's like pretty deep in this, like in in the stack.
Uh there's a lot of uh a lot of great tech that can be built on top of that.
Yeah.
Um it's also just part of a security solution.
I mean, I uh there's no way I'm gonna run a coding agent on my machine and like, you know, with dangerously skip permissions, for instance.
I'm gonna do that in a sandbox.
No.
No, I'm doing it.
I'm doing oh my gosh, I'm doing it.
I'm not doing it locally.
I I I have a sandbox.
I have a sandbox in a cloud provider that can't get out.
And I'm you know, I don't put I don't put my SSH keys on that thing.
I just I push things to it, you know.
And I think there's I know my colleague Avery, our CEO, he's been experimenting a lot with this kind of stuff too.
Of just like, yeah, because like velocity is super important, you know.
These agents nowadays, and like just like the code, like LMs are just improving dramatically week over week.
I want to give them I want to really leverage that safely.
And I think aperture is part of that safety solution, but it's not the whole answer.
So you built in the AI gateway with identity baked in, yep.
Uh API keys not have to be passed around.
There's a lot of benefits here.
You mentioned the desire for partners.
I mean, give me an idea, give me an idea of like what you mean by that.
You said we're not gonna build all that, and you alluded to all that.
We talked through a bunch of stuff.
I mentioned agent policy and stuff like that.
What do you mean by all of that?
And how can partners step in?
How can an individual, a team of one, step in and become a partner of of Tail Scale and build out what you're not gonna build out?
Well, um, let me think.
I mean, they can contact me.
Uh happy to have a conversation with companies and individuals working in this space.
What's the best way to reach you?
You want to you want to say your email address here, or is it is there another way to get a hold of you?
Oh, just aperture at tailscale.com.
Okay.
I'll see it.
Yeah.
That's you.
Uh yeah.
Well, it's it's it's me and the team.
It's it's you plus, yeah.
It's the proverbial you.
Yep.
And uh aperture is A-P-E-R, is that right?
Or A-P-U-R.
E R.
Yeah.
I always misspell.
I mean, listen, ever since Aperture, Apple had this software called Aperture for photographers, if you know this, you know, 15 years ago.
It was amazing.
They don't make it anymore.
I've never been able to spell aperture properly.
It's like I gotta remind myself, uh, you know, I have I before e after like I'm every single time my brain doesn't get it okay.
So please spell aperture if you don't mind info.
Aperture at tailscale.com.
Yeah, A-P-E-R-T-U-R-E.
Pull that in the show notes for everyone to get a hold of you.
Because I'm really curious about this.
I think this is cool.
Uh, I mean, when I first heard about it, I was thinking, oh my gosh, big brother.
But then we kind of need a big brother in a way, because agents can't be trusted.
Until we could trust them, we sort of have to big brother them because they're already big brothering us in a way.
And logging all the things to me gives me some peace of mind because I want history.
I want my own history.
And from uh an enterprise standpoint and maybe a peer standpoint, and it might be a little weird to see what David's coding or how you're prompting, but I don't know.
I think we're getting it gonna have to be just okay with that to some degree.
But I'm gonna want to learn or peek over your shoulder and say, well, David's clearly like 10Xy my own 10x here.
What is he prompting here?
You know, how is he what is this magic he is he is doing here?
Maybe it's just like do it.
Yeah, is that your prompt, David?
Just do it.
Oh, there's been I've been tempted.
Sometimes that's my prompt.
It's like, yeah, presents an idea, do it.
Yeah, I like that.
Sounds great.
That's an amazing idea.
I gave you the original idea, you morphed it.
Now it's amazing.
Yeah, do it.
Do plan one.
I love that.
Yeah.
I I spent a lot of time in planning mode with Claude.
Yeah.
Yeah.
Uh it it is crazy though, like how fast those tools can let you move uh and the stuff they can do if left unchecked.
You gotta be careful.
Um I I think, you know, I guess comments on a couple of things you just said.
Like there's there is uh we've got some basic permissions with Aperture right now, you know, like users can only see their own stuff, for instance, admin can it's admins can see the world.
Like we've got a lot of stuff to implement there about, oh, maybe we only want a team to be able to see like you know, logs within their team.
Maybe we only want a manager to see if the team members like there's all these different kinds of I guess uh access models that we need to explore with customers on this.
And we are doing that.
So there's um there's a lot of room for new features there.
But again, we're just trying to keep sip things simple at this point.
Uh in terms of one of the neat hacks that we've done internally, and I really like this one, is that you can actually point a coding agent uh, because we have an API.
There's like there's a there's like a whole bunch of endpoints inside of Aperture.
And one of the endpoints inside of your tail net is that you could actually get your own logs out of it.
Right.
So you can you could point a coding agent and can say, like, oh, I want you to explore like basically how you've worked in the past.
Right.
And so you you can start to get like these recursive, I guess, like learning or feedback loops with a coding agent like reviewing how it's worked in the past or reviewing its its previous logs.
Right.
And I think you can unlock, I think there's a lot to unlock with that.
You know, we've just only started to scratch the surface, but it's yielded some really interesting insights for us as we've been digging into like, oh, how do these protocols work, for instance?
Like how do these coding agents tend to function?
Like how like when they start delegating stuff or using subagents, like how does that kind of how do those mechanics work?
And so I think for a home labber that you if you want to explore more about just like the protocols and the request headers and the bodies and like how these coding agents are sort of like you know, how the API calls evolve and how the contexts evolve, it's super neat.
I like the fact that you have access models.
And that gives me more peace of mind because you know it's gateways are great.
You already have it at the network, so you can't hide from that.
And uh IP addresses, DNS, those reveal a lot about an individual.
Uh but you know, an LLM and your interaction with a model like we do today reveals, you know, a lot more because you may say, Hey, I'm new to this scenario, and like maybe your team thinks you're really steeped in it or whatever.
I think you're more advanced.
And that's not so much judgment happens, but you you may show or have to be more truthful with your awareness and level of understanding of something.
And that may be either one be embarrassing or a fireball offense, or I don't know.
But then you start to think about like, okay, who has access to that information?
I realize this is all work stuff, but I have to be, and we're almost getting more and more.
I don't want to say the word intimate, but it's it kind of is the word, a little bit more relational, a little bit more intimate with the other side.
Let's just say the the machine behind the machine.
Uh, because it feels a lot like we're talking to something, a peer.
And it in a lot of ways, it acts as a peer, as an educational peer.
Education is dramatically changing, let alone uh code level understanding.
Like navigating code base is so much different today than it was like like not even 180, like like 720 difference uh in terms of like spins that it's just dramatically different to navigate a code base and have an understanding in a world where you have agentic properties available to you.
And in a lot of cases, you might be more forthcoming with the agent, because you have to be, you know, to get it to give you what you need, you kind of have to be like, you know what?
I don't know a lot about this subject matter.
You know, can you, can you, can you steep me in it?
Can you school me?
Where what do I need to learn here?
Can we build a toy application?
And then you get launched into it.
But it can be a little revealing.
Uh sometimes maybe too exposed.
And that's where I that's where I was like, you know what?
I like the fact that you have guardrails because I was a little a little apprehensive myself.
In my own self-hosted version of it, hey, that's cool.
Because I'm not judging me.
But somebody else might if I'm in an enterprise.
Oh, totally.
No, I I I have like when I started using coding agents, like the prompts are like I I almost I'm scared to go back because it might be in right like it's like, oh I was burning a lot of tokens on useless things.
Like I didn't know how to use this tool properly.
But the thing is we're all trying to learn how to use these tools properly.
And the thing is they change every like few months.
Like I think back to you know what like like earlier sonnet versions back in September versus what I'm using now.
Yeah.
It's night and day the capabilities.
Like you have to like teams really have to stay on top of this stuff if you want to keep up with it.
Yeah.
So I I you know yeah privacy obviously like uh security is key.
I think I think people have to be pretty open and candid and transparent if they actually want to learn with this space and like really leveraging and getting advice from other people too which is another one of the reasons I wanted I'm so excited about Aperture is just in terms of like the transformation I think it can help our company make because you know we're seven years old.
You know we've got a lot of incredible engineers and other people working on all sorts of technical stuff.
When we started the company LOMs basically didn't exist you know and look where the industry is now like I dare think where it's going to be a year from now we have to adapt and learn more about these things.
But a world where people are just like just clicking yes all the time is not a world I want.
I I want us to be like like leveling up and I think to do that you have to you have to sort of look at what you did.
It's kind of like code reviews just much more aggressively it's like prompt reviews and and just understanding like how do these tools work?
How do our workloads needs to change like where did this like where did this agent go wrong like where did it make the wrong kind of assumptions how do we have to shape our code base so it doesn't do this in the future that kind of stuff like there's there's so much uh I guess knowledge that we have to uncover of how to work with these tools going forward to make them effective I love the fact that uh prompt review and uh PR uh pull request just naturally translate you know I don't know that was like uh serendipitous 17 years ago when GitHub and set that itself and said okay pull requests are the thing fork yourself kind of stuff yeah uh that was like the the the the the founding uh detail of GitHub was being able to fork a code base and submit a pull request it was brand new territory and and now look at us we're just talking to our code bases I still can't believe it still can't believe it I have to pinch myself every day like that's crazy.
I don't know about you but um uh GPT 5.3 codex is really I've just never worked with an agent that was that um advanced it's like the working with the most advanced engineer ever.
And in some cases I'm a little pissed because it's like it knows way more than I think I would ever know.
And I'm I'm almost like yeah I know a lot about this but like wow you clearly you're speaking a whole different language and you're moving way faster than I can ever do it.
And it it's a little scary it's a little scary but I I think this gateway idea has got some merit especially in the field you're trying to go to which you need a killer app right you need a killer app on on tail scale that is beyond the VPN right so this is taking you beyond the VPN it's taking identity in a whole new place it's securing our API keys in ways we've never been able to before in a world that is burgeoning and very tumultuous in terms of security so I I'm all for this my only desire would be uh self-hosted at some point and whenever you get to that obviously you know I'll I'll check it out in a metaway between now and then but I'm gonna be in a world where and this is just my own personal opinion I've just been in this world where these are the kind of things that I want to have sovereignty over and I already have compute so why not dedicate my own compute to it I'm happy to pay a licensing fee or you know the business model however it needs to go but I'm in this world where I want to self-host a lot more than I ever wanted to before and it's not it's not that I don't trust the world it's what I want to have more control over the world I'm building and they're already interconnected I've already used uh tails I already have my tail net I haven't tapped into multi-tail nets yet, but I'm going to.
That's where I'm at is is uh give me the self-hosted version of it because that's gonna be that's gonna be fun.
I hear you.
Yeah, I hear you.
What's uh what's left?
What I what have I not asked you, David, about your journey with tail scale, what you think people do and don't know.
Like, what is the biggest myth?
If you could debunk a myth about tail scale, what what might that be?
Do you often have to debunk myths about what you do and what you don't do and how deep of a well you all have?
Uh well one of the one of the myths is that we're just for home labers or small teams.
That's not true.
Like we've got a significant number of enterprise customers.
Uh but I've often heard and called, or like, you know, we'll bump with the people at conferences who are like, oh, like so and so was saying that you know, you're just for home labs or you're just for small teams, you're just like a hobbyist kind of thing.
It's like, no, we're we're serious.
Like, you know, having a free motion.
Some of the most advanced engineers I've ever known of.
Like you got some really talented people in your team.
I'm I'm very much like you really do.
You have some serious engineering talent.
Yeah, we do.
Uh yeah, I'm very grateful for that.
It's yeah, it's an incredible team.
Um Brad Fitzpatrick is one of the ones I'm thinking of.
Like, he's been on our podcast, Go Time.
We used to host this podcast called Go Time.
That's why I'm so emphatic about GoForCon, even too, and and go the language.
And Brad's been on that podcast.
And I think I'm not sure who all is still there over these all these years, but a lot of folks that I've just paid attention to and have like leaned on for wisdom on how to morally navigate the software we're building, uh, even from a technical level, like just bar none, some really awesome people there.
Yeah.
Yeah.
I I I count my blessings uh with the with the team we have.
Um, but yeah, to to talk about the myths.
You know, we are we're never gonna give up having a free plan.
You know, I think free forever.
Yeah, free forever.
Like we're gonna have, we're always gonna have something, and we're gonna be pushing more and more into that over time.
Uh, that's really important to Avery and me and so many other members of the of Tailscale in general.
Uh, but you know, we are building more and more stuff all the time, and a lot of that has uh enabled us to just take on bigger and bigger and bigger customers to help, you know, frankly pay the bills and help us grow and help us expand.
And it just makes Tailscale a better product for everyone, right?
And so uh we are definitely like enterprise ready.
And that has been one of the myths that over the past couple of years we've we've done a lot to, I think, to establish that, but there's always more, I think we could do because we've come bottom up.
And so there's always that, you know, people who learn about you in the early days, that's how they think about you.
It's like the curse of SAS.
Like once somebody adopts you, you have to spend so much time re-educating them because like you know, the products change and evolve a lot, right?
And so we had a lot of early adoption.
Um, and we have to go back and just talk more about what we've done, the technologies we build and like where we're acting, the kinds of customers and stuff.
Uh that's one of the myths.
Uh the other one is that we're just a VPN.
Tailscale's more than a VPN.
A lot of like conventional VPNs, it's all about IPs and connections.
Like Tailscale bakes identity in.
Um it's yeah, it it's a fundamental guarantee of every connection.
Like you know who and what is connecting.
Like if it if it can if it can connect to you, it's already authorized, and you know you can tell who it is immediately.
Like you know who it is.
Um, and that is that's more of a paradigm shift where people think about like connectivity and identity being the sort of the same thing or paired very closely together, which is uh not how most people think about like networking.
They think about oh, I've got this connect, I've got this connection, and then I've got identity that's like way on top somewhere, like layer seven.
It's like, oh no, no.
But tail scale is like it's baked in.
And if it is, you can just do so much more with that.
You don't have to think it or worry about identity.
And I think that once people get their head around it, like you can see these lights go off.
Um that's that's the real sort of yeah, yeah, like I said, paradigm shift I want to bring to people.
I do not know what you can do differently because I'm a I would say I'm a pretty steep user of Tailscale.
And and even some of the things you're saying now about these myths, they're myths, even to me as a as a daily active user of Tailscale.
I don't know what you do or what you can do to solve for that problem.
Um, but I agree, not just a VPN.
It's unclear to me how in all the ways that my identity is attached to my it obviously makes sense not that you explain it like that, but it's just not obvious in my daily use of it.
Yeah, uh, even when I uh SSH around my network, I I don't do it via Tailscale.
I don't think I will SSH via Tail Scale with like maybe let's say if I have a machine that has a host name of Cineplex.
So my Plex machine is called Cineplex.
So I like it like that.
I'll just SSH Cineplex.
I don't know if I'm using any special niceties of Talescope besides maybe host name mapping.
That's about it, probably.
I'm not using my tail scale identity, I don't think.
Uh that's that's maybe, and that's maybe my own fault, maybe it's your fault.
I don't know.
But uh I agree.
Like I feel like I learn more and more about what Tail Scale can do for me because you do so much.
And I I don't know how you explain to folks more differently than you already do.
Um besides just keep I guess keep trying.
I don't know.
It's yeah, it's true.
I mean, it's it's um yeah, good infrastructure gets out of your way really quickly.
And that's been a guiding principle for us ever since day one.
Yeah, you know, so it's you know, you don't want infrastructure to make noise.
Uh at the same time, it would be awfully nice if more people sort of I don't know, like just understood all the cool stuff that we were building.
So it's it's a very tricky balance because we we try to be very sort of quiet, just works like always works in the background.
Do we want to make a lot of noise about a new feature?
Like it it it yeah, it creates tension, I'd say.
We try to be very cautious with that.
Like, even with Aperture, I mean, just thinking out loud here, I think you know, one of the ways you can show off a lot about it is not so much build to applications, but show off the cool things you can do with it.
Like this podcast is one example, diving into it.
There's just so much you could do with an AI gateway, which I think is worth exploring.
Um, that the way you can you can explain things to folks is really just to show it off, demo it.
And maybe that's conferences, maybe that's in the hallway track, maybe it's via YouTube and Alex's team and what they're doing there.
But I think there's so much you can do with identity on your network that I'm just now thinking about that I that I personally care deeply about on a daily basis.
Cause I'm building things that require it, and I'm not leveraging any of this tooling.
I'm doing it the hard way, despite being a user.
And it I'm almost angry at you and and the proverbial you for not, and maybe you do.
And maybe Alex is like, Adam, just watch my YouTube.
And then maybe that's the easy button there.
Alex, but like there's a lot of content.
Yeah.
Right.
That's the thing that's just a lot of things.
Yeah, there's a lot of things.
But you know, um, but yeah, like don't don't put an LLM on a public port.
You know, like don't don't don't put an LM like on the public internet.
Don't do that, right?
But you can you can share that with like you can create a private network and you can share that with your friends with tail scale.
Like there's like so many little things like it's like, oh, I just I didn't realize there's sort of an easier, better, more secure way.
I'm telling you, I need to know the easier, better, more secure way, the tail scale way, the tail net way, the multi-tail net way.
I can't wait to play.
Oh my gosh, I punned and I rhymed at the same time.
Uh David, it's been so awesome talking to you, going into the details of this.
I'm a big fan, as you already know.
Uh I'm looking forward to the self-hosted version of it, uh, as you already know, because I've said that a couple times already.
Yeah, yeah.
I think that's where the future is at in a lot of cases here.
I can see what you're spinning up instances to to achieve velocity, but I think ultimately sovereignty is the key uh to my book at least.
Um there you go for that front there.
Anything left in closing, what else you want to say?
Aperture at tail scale.com.
We'll put that in the show notes, of course, to reach out to you to become a partner or a builder who's got some ideas on top of TSNet and the things you're building there, or maybe even aperture itself.
What else?
I invite people to reach out, whether it's like for partnerships or ideas or feature requests or what like I'm I'm here, I'm accessible.
Like I don't want people to think that you know, just because tail scale is a few years in and we're sort of the size we're at right now, that like I'm unreachable and that we're you know we we definitely don't have all the best ideas.
We've got some good ones, uh, but we want to work with a lot of other people on and help them get their ideas to market like quickly and in a more safe, secure, and uh expedient way.
So yeah, please like reach out.
I'd love to hear from people.
Maybe some office hours for for you.
Maybe you could do some office hours.
Would you entertain that?
Yep, yeah.
No, let's come up.
That's that's definitely uh, I think once we get through a bit more of this push on aperture, record it too.
I'd I'd turn it into some content because I mean like I would I would pay attention to the behind the scenes of that.
I think that's like peer-to-peer is where I think a lot of developer, like that's one of the ways we we really educate folks that uh listen to our pod, and you know, we have a lot of sponsors who sponsor our stuff.
But one of the directions we take is is not just throwing ad out there, but something that's like it goes behind the scenes, it's it's informative and it's peer-led.
Yeah, and you can kind of see okay, well, this one team over here's got this idea for how they can leverage aperture or an API net or sorry, an AI net, uh an AI gateway.
Um, and now I can see what they're using it for, and now I've got some ideas as well, kind of thing.
Yes.
I think office hours could be kind of fun.
And to speak to the top like you are and your team members, that'd be kind of cool to like bring some questions, dig into how that works out, throw some ideas out there, let it be a little loose, but also a little structured, and uh yeah, that'd be cool.
I could see that happening.
Yeah, it'd be a lot of fun.
I'm uh yeah, I just I love helping people get their ideas to market and removing some of the pain.
Like it's just it's such a fantastic experience.
And uh I just want to see and help with more of that.
So yeah, very, very happy to talk to people about their ideas and work on collaborations and partnerships and stuff like that.
All right.
Well, David, thank you so much.
Appreciate you.
Awesome meeting you awesome conversation.
Yeah, thank you.
We'll see you again soon.
Well, that's it.
The show's done.
Thank you for tuning in.
Big thank you for being a listener of this podcast.
If you haven't yet, become a member.
It is free.
Yeah, you can go to change law.com slash community, free to join.
Hang with us in Zulup chat.
Everyone's there.
Everyone's welcome, and you are welcome.
And I want to see you there.
And if you love this show, just a little bit more than you love every other show out there, and you want to go deeper, get bonus content, get closer to the medal, drop the ads, support the show.
We have a membership that is not free.
It's called Changelog Plus Plus.
Learn more at Changelaw.com slash plus plus.
It's better.
It is better.
You know why it's better?
Because you get bonus content, you get the little extras, you get close to that medal, and like I said, you support the show.
Big thank you to our sponsor for this show today.
Big thank you to BMC for being our beats freak in residence, the Breakmaster Cylinder.
My gosh, those are awesome beats.
And thank you to you for tuning in to this show.
That's it.
We'll see you again soon.