<!-- mobian-agent-page publisher="dailydev" canonical="https://daily.dev/posts/how-to-generate-mergeable-code-with-a-context-engine-peter-werry-unblocked-j9pgagvdm" -->

---
title: How to Generate Mergeable Code with a Context Engine —...
description: A conference talk from Unblocked's Peter Werry pitches the concept of a 'context engine' as the missing layer for AI coding agents. The argument is that agents...
canonical: https://daily.dev/posts/how-to-generate-mergeable-code-with-a-context-engine-peter-werry-unblocked-j9pgagvdm
twitter:card: summary_large_image
twitter:site: @dailydotdev
og:type: website
og:site_name: daily.dev
og:title: How to Generate Mergeable Code with a Context Engine — Peter Werry, Unblocked | daily.dev
og:description: A conference talk from Unblocked's Peter Werry pitches the concept of a 'context engine' as the missing layer for AI coding agents. The argument is that agents...
og:url: https://daily.dev/posts/how-to-generate-mergeable-code-with-a-context-engine-peter-werry-unblocked-j9pgagvdm
og:image: https://api.daily.dev/og/posts/J9pgAGVdm.png
og:image:alt: How to Generate Mergeable Code with a Context Engine — Peter Werry, Unblocked
og:image:width: 1200
og:image:height: 630
og:locale: en
---

> ## Documentation Index
> Fetch the complete documentation index at: https://daily.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# How to Generate Mergeable Code with a Context Engine — Peter Werry, Unblocked

**[AI Engineer](https://daily.dev/sources/aidotengineer)** · 18 min read · 1 upvotes · 0 comments

## Summary

A conference talk from Unblocked's Peter Werry pitches the concept of a 'context engine' as the missing layer for AI coding agents. The argument is that agents reset their knowledge on every task, unlike human engineers who build tribal knowledge over time, and that dumping wikis or entire codebases into context windows doesn't solve this because agents suffer from 'satisfaction of search' and fail to distill understanding of how systems fit together. A live demo shows Unblocked answering architecture questions with generated diagrams, participating in Slack discussions, and feeding richer context into Claude Code to produce better optimization plans faster and cheaper than without it. Additional demos cover a code-review agent that surfaces past senior engineer feedback and an autonomous PR-generation agent that ties fixes back to Slack conversations and prior decisions. The talk closes with mentions of two open-source projects (a document query engine and an engineering social graph) and a customer quote citing 50% fewer tokens and faster triage.

## Full article

daily.dev links to this article rather than hosting it. Read it at the original source: <https://www.youtube.com/watch?v=qdAkxLoYNI8>

## Questions this post answers

### Why do AI coding agents struggle even with a large context window and a full codebase dump?

Agents get distracted rather than helped when an entire codebase and architecture docs are stuffed into a context window, even one sized at a million tokens. There is simply more organizational context than can fit, and irrelevant material pulls the agent's attention away from the task, wasting tokens and time instead of improving reasoning.

_Anyone tuning agent context strategies can compare approaches and pitfalls surfaced across daily.dev._

### What is 'satisfaction of search' and how does it apply to AI coding agents using a wiki for context?

Satisfaction of search is a radiology term for stopping after finding one apparent answer and missing other important indicators nearby. Coding agents exhibit the same pattern: given access to a wiki, they can search and find something that looks correct, then stop, without understanding whether it's the most relevant or complete context for the task.

_Developers debugging why agents miss context can dig into failure patterns like this on daily.dev._

## Similar posts on daily.dev

- [Want better AI outputs? Try context engineering.](https://daily.dev/posts/want-better-ai-outputs-try-context-engineering--l4fmqqyz5) · GitHub Blog · 1 upvotes · 0 comments
- [The Context Tax of Agentic Development](https://daily.dev/posts/the-context-tax-of-agentic-development-spfv0nbhr) · Expedia Group Tech · 0 upvotes · 0 comments
- [Context Engineering for Coding Agents](https://daily.dev/posts/context-engineering-for-coding-agents-ilil6pcpq) · Martin Fowler · 2 upvotes · 0 comments

---

Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#mcp](https://daily.dev/tags/mcp), [#claude-code](https://daily.dev/tags/claude-code), [#code-review](https://daily.dev/tags/code-review)

[View this post on daily.dev](https://daily.dev/posts/how-to-generate-mergeable-code-with-a-context-engine-peter-werry-unblocked-j9pgagvdm)

```json
{"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]}
{"@context":"https://schema.org","@type":"TechArticle","headline":"How to Generate Mergeable Code with a Context Engine — Peter Werry, Unblocked","url":"https://daily.dev/posts/how-to-generate-mergeable-code-with-a-context-engine-peter-werry-unblocked-j9pgagvdm","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/how-to-generate-mergeable-code-with-a-context-engine-peter-werry-unblocked-j9pgagvdm"},"datePublished":"2026-08-27T18:30:26.771Z","dateModified":"2026-09-14T07:56:32.234Z","description":"A conference talk from Unblocked's Peter Werry pitches the concept of a 'context engine' as the missing layer for AI coding agents. The argument is that agents...","image":"https://i.ytimg.com/vi/qdAkxLoYNI8/sddefault.jpg","thumbnailUrl":"https://i.ytimg.com/vi/qdAkxLoYNI8/sddefault.jpg","isAccessibleForFree":true,"articleSection":"AI Engineer","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"AI Engineer","logo":"https://media.daily.dev/image/upload/s--u5PucxNT--/f_auto/v1724338940/logos/aidotengineer","url":"https://daily.dev/sources/aidotengineer"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/how-to-generate-mergeable-code-with-a-context-engine-peter-werry-unblocked-j9pgagvdm","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,mcp,claude-code,code-review","timeRequired":"PT18M","video":{"@type":"VideoObject","name":"How to Generate Mergeable Code with a Context Engine — Peter Werry, Unblocked","description":"A conference talk from Unblocked's Peter Werry pitches the concept of a 'context engine' as the missing layer for AI coding agents. The argument is that agents...","thumbnailUrl":"https://i.ytimg.com/vi/qdAkxLoYNI8/sddefault.jpg","uploadDate":"2026-08-27T18:30:26.771Z","duration":"PT18M","url":"https://api.daily.dev/r/J9pgAGVdm","embedUrl":"https://www.youtube.com/embed/qdAkxLoYNI8"}}
{"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"AI Engineer","item":"https://daily.dev/sources/aidotengineer"},{"@type":"ListItem","position":3,"name":"How to Generate Mergeable Code with a Context Engine — Peter Werry, Unblocked"}]}
{"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/how-to-generate-mergeable-code-with-a-context-engine-peter-werry-unblocked-j9pgagvdm#faq","mainEntity":[{"@type":"Question","name":"Why do AI coding agents struggle even with a large context window and a full codebase dump?","acceptedAnswer":{"@type":"Answer","text":"Agents get distracted rather than helped when an entire codebase and architecture docs are stuffed into a context window, even one sized at a million tokens. There is simply more organizational context than can fit, and irrelevant material pulls the agent's attention away from the task, wasting tokens and time instead of improving reasoning. Anyone tuning agent context strategies can compare approaches and pitfalls surfaced across daily.dev."}},{"@type":"Question","name":"What is 'satisfaction of search' and how does it apply to AI coding agents using a wiki for context?","acceptedAnswer":{"@type":"Answer","text":"Satisfaction of search is a radiology term for stopping after finding one apparent answer and missing other important indicators nearby. Coding agents exhibit the same pattern: given access to a wiki, they can search and find something that looks correct, then stop, without understanding whether it's the most relevant or complete context for the task. Developers debugging why agents miss context can dig into failure patterns like this on daily.dev."}}]}
```

