← back
Context Is the New Code — Patrick Debois, Tessl
Takeaway
Context artifacts (prompts, skills, agent.md) need the same generation, testing, observability, and versioning rigor we apply to code.
Summary
- Patrick Debois (DevOps originator) proposes a 'context development lifecycle' analogous to SDLC: generate, test, distribute, observe, adapt context as we now generate code.
- Reusable prompts/skills/agent.md files become assets; pull library docs, MCP-fed Slack/GitHub context, and spec-driven planning into the generation step.
- Advocates evals for context like linting (skill description length) and 'Grammarly for prompts' to catch underspecified context; LLM-as-judge tests checking conventions (e.g. every endpoint prefixed /awesome/).
- Adds judge-with-tools end-to-end tests that actually curl the running endpoint, treating context changes like code commits with full CI.
context-engineeringevalsai-coding
Original description
As AI coding agents become more capable, context is starting to matter as much as code. Yet while code has version control, review, testing, CI/CD, and production observability, the prompts, rules, and memory that drive agents are still often managed like ad hoc hacks. Patrick argues that context needs its own engineering discipline. He introduces the Context Development Lifecycle: Generate, Evaluate, Distribute, and Observe, along with the team practices that make context a shared, repeatable, and improvable part of software delivery. The session also explores the larger context flywheel, where better context leads to better agent output, which creates better observations, which in turn improves context again. Speaker info: - https://x.com/patrickdebois - https://www.linkedin.com/in/patrickdebois/ Timestamps: 0:00 - Introduction to the talk 1:14 - Why context is the new code 2:37 - Introducing the Context Development Lifecycle 3:50 - Generate: Creating context for agents 6:26 - Evaluate: Testing your context 13:59 - Distribute: Sharing and packaging context 17:49 - Observe: Monitoring and feedback loops 22:33 - Conclusion and the context flywheel 24:49 - Q&A session