EP 4 - The Build - OpenAI MCP, AR Agents, Open Agent Platform and more

In this episode of The Build, Cameron Rohn and Tom Spencer discuss OpenAI MCP, AR Agents, Open Agent Platform and more, mapping technical trade-offs and entrepreneurial opportunities. They begin by unpacking AI development and tools, with concrete references to Langsmith, MCP tools, Codex, and the Agent Development Tool while highlighting API integration patterns and Sync functionality exploration. The conversation then shifts to technical architecture decisions, comparing agent orchestration, webhook strategies, and hosting options like Vercel and Supabase for low-latency agent state and developer workflows. They explore building in public strategies next, debating Stock vs Impactful Content, metrics for community-driven roadmaps, and Creative Tooling Follow-Up as a monetization path. Later, the discussion covers platform integrations—ChatGPT vs Anthropic Integrations, and experiments with Google X XR Platform for AR agents—drawing practical implications for system design, observability, and testing. Throughout, entrepreneurship insights surface: founder priorities for developer tooling, go-to-market approaches for AI products, and tech-driven facility design as a niche business idea. Finally, they synthesize lessons on developer ergonomics, open source community building, and iterative architecture choices, leaving listeners with a forward-looking takeaway: prioritize modular, observable agent infra and public iteration to accelerate product-market fit.