AI coding tools push startups toward non-technical buyers
Snyder argues that AI coding tools make DIY software viable for many users, forcing startups to target customers who cannot code themselves. This shift is critical for maintaining value against general-purpose models.
Snyder discusses the impact of AI coding tools, specifically referencing 'Claude Code' and 'vibe coding,' on the startup landscape. He notes that while AI makes building software cheaper and faster, it increases the difficulty of finding demand because 'do it myself' becomes a viable option for many users. To succeed in this environment, Snyder advises startups to target customers for whom DIY coding is not an option, such as insurance or logistics firms, or those who have attempted DIY and hit technical walls.
He cited a friend's company that found customers who tried to build their own 'company brains' using tools associated with Karpathy but failed when trying to integrate team data due to privacy and complexity issues. Snyder also addresses the sustainability of AI-native companies like Harvey and Lovable. He suggests that while models improve, the 'last mile' of product experience and specific customer needs allow these companies to maintain value despite the threat of commoditization by general-purpose models. Snyder posits that the rise of AI coding tools forces startups to focus on customers who cannot or will not code themselves, as the DIY option erodes the market for simple software solutions.