Context Is the New Code
summary
Argues that context needs its own discipline and gives a concrete Context Development Lifecycle: Generate, Evaluate, Distribute, Observe. Shows reusable formats like agent.md / Claude.md and skill packages in registries, demonstrates tests called evals (run e g five times to measure nondeterministic success), proposes error budgets, describes sandboxed judges, agent logs and an AI SBOM for security, and explains packaging, CI/CD integration, and a /awesome endpoint example to validate distributed context.
AI-written summary · about
chapters (9)
- 0:00Introduction to the talk
- 1:14Why context is the new code
- 2:37Introducing the Context Development Lifecycle
- 3:50Generate: Creating context for agents
- 6:26Evaluate: Testing your context
- 13:59Distribute: Sharing and packaging context
- 17:49Observe: Monitoring and feedback loops
- 22:33Conclusion and the context flywheel
- 24:49Q&A session
related talks
28mContext Is the New Code
15mContext Is the New Code for AI Agents
27mContext is the new code: The context development lifecycle- 27mContext is the new code: The context development lifecycle
37mThe Context Flywheel for AI Coding Teams
15mUnderstanding Context Engineering for Agentic Systems
31mBuilding the Context Engine AI Agents Need
35mFrom Prompt Engineering to Context Engineering: Lessons from an AI Code Review Product