Peter Werry compares coding agentsAn AI coding agent is a tool-using AI system that can inspect, modify, and validate software within a repository. with employees who are repeatedly onboarded without enough organizational memoryInstitutional knowledge is the accumulated decisions, practices, context and experience that an organization relies on to work effectively.. Repository instructionsRepository context is the relevant source, configuration, history, documentation, dependencies, and worktree state needed to understand a software task., skills and MCP tools can help, but an agent may still stop at the first plausible answer because it does not know which architectural decisions, prior discussions or future plans matter to the task.
He describes a context engine that connects code with pull requests, Slack conversations, documents and team expertiseAn AI context engine gathers, reconciles and ranks relevant information so an AI agent can act with the history, decisions and permissions surrounding a task.. In a live planning comparison, the context-aware agent found proposals and architecture details that the baseline plan missed, producing a more specific implementation path with fewer investigative steps.
Werry also shows how the same organizational contextContext engineering designs the information, instructions, memory, and tool state an AI receives so it can perform a task reliably. can support code review and regression diagnosis. One example links a behavioral change after a model switch to earlier team discussion and then generates a corrective pull request. Promotional calls to sign up, scan a simulator code and attend another talk are omitted.
Watch on YouTube
