Burry's Short on AI Giants

In this episode of The Build, Cameron Rohn and Tom Spencer discuss Michael Burry's short on AI giants and extract lessons for AI development and startups as they connect market signals to engineering choices. They begin by unpacking the headline — Burry's positions against Nvidia and Palantir — and trace the implications for AI agent development, highlighting concrete tools like Langsmith and MCP tools for orchestrating multi-step agents and testing emergent behaviors. The conversation then shifts to developer tools and workflows, where Vercel and Supabase appear as exemplar platforms for rapid prototyping, CI/CD, edge deployment, and low-friction backend services that accelerate building in public. They explore technical architecture decisions next, weighing serverless vs. containerized approaches, vector store selection, orchestration patterns, and trade-offs in observability and latency when composing models and agents. Throughout, they layer entrepreneurship insights about monetization, community-driven growth, and transparent roadmaps — practical tactics for founders to publish progress, collect feedback, and iterate. The episode synthesizes developer-focused takeaways on tooling, architecture, and public-facing product strategies. They close with a forward-looking call: founders and engineers should keep building in public, instrumenting systems with the right stack, and iterating quickly to capture emergent AI opportunities.