Mergeable by default: Building the context engine to save time and tokens — Peter Werry, Unblocked
Source
youtube.com
Author
AI Engineer
Date
Context
Code can be technically plausible yet conflict with a team's conventions or earlier decisions. This talk's proposed context engine addresses that gap by selecting organizational information for the specific coding task and person asking.
The published description identifies conflicting sources, access permissions and task relevance as central design problems. Resolving those questions determines which information an AI agentAn AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.Full definition → should receive before writing code. The speaker's stated aim is to avoid paying twice for wrong-context windowThe maximum amount of text a model can consider at once — its working memory for the current conversation or task.Full definition → code, once in model usage and again in lengthy review.
Terms in this piece · Glossary
context window — The maximum amount of text a model can consider at once — its working memory for the current conversation or task.
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
videoHow to Write Your First AI Eval - SallyAnn DeLucia, Arize AI
videoBuild Your Own Team of Managed Agents - Jack Wotherspoon, Google DeepMind
videoHow to Build an AI Code Review Agent - Daksh Gupta, Greptile
videoHow to Build Voice Agents That Don't Fall Apart When You Interrupt - Jesse Hall, LiveKit
