AI in Options Trading: A New Approach

In this episode of The Build, Cameron Rohn and Tom Spencer explore AI-driven approaches to options trading, emphasizing how agent architectures and modern developer tooling reshape both product and process. They begin by breaking down the architecture of AI agents for trading, discussing state management, orchestration, and evaluation patterns while referencing Langsmith for experiment tracking and MCP tools for monitoring and policy control. The conversation then shifts to developer workflows, where Vercel and Supabase surface as deployment and data-layer examples that simplify CI/CD, edge functions, and realtime data needs for low-latency trading systems. They explore building-in-public strategies and entrepreneurial trade-offs, weighing open source community gains against monetization paths such as hosted platforms, API tiers, and consulting. They examine concrete technical decisions—model selection, vector stores, observability, and latency optimization—and practical integrations with Langsmith pipelines and Supabase backends to support iterative research and reproducible results. Along the way they address UX for decisioning agents, risk controls, and regulatory considerations that influence architecture and go-to-market choices. The episode closes with actionable guidance: prioritize modular agent architecture, instrument experiments with tools like Langsmith and MCP, and use public builds and transparent roadmaps to accelerate product-market fit and community momentum for developer-led startups.