Context is the new code: The context development lifecycle
summary
Claim: promptable context becomes the primary artifact and must be engineered like code with a Generate→Evaluate→Distribute→Observe lifecycle. The talk shows concrete tools and patterns: an agent.md file for reusable prompts, skills registry packaging for reuse and versioning, evals that run LLM judges and even curl checks to validate endpoints, and harnesses/loop engineering that add CI/CD, linting, security scans, telemetry and sandboxes so agents actually improve rather than drift.
AI-written summary · about
chapters (18)
- 0:00Context as the new code
- 1:14DevOps to context-driven dev
- 2:42Reusable context generation
- 4:31Testing context with evals
- 6:41Advanced context evaluation
- 8:28Distributing context via registries
- 10:35Governing context packages
- 12:20Observing context performance
- 14:45Optimizing context with harnesses
- 16:51Loop engineering systems
- 18:17Autonomous context automation
- 20:00Organizational harness building
- 21:50Maturity: code to loop engineering
- 22:30Context workflow guardrails
- 24:14Review interfaces for context
- 25:15Context flywheel and platform layers
- 26:20New developer roles
- 26:41Closing and collaboration
related talks
27mContext is the new code: The context development lifecycle
28mContext Is the New Code
27mContext Is the New Code
15mContext Is the New Code for AI Agents
37mThe Context Flywheel for AI Coding Teams
35mFrom Prompt Engineering to Context Engineering: Lessons from an AI Code Review Product
15mUnderstanding Context Engineering for Agentic Systems
31mBuilding the Context Engine AI Agents Need