The Data Context Layer for Agents — Yoni Michael & Brandon Callender, Typedef
Source
youtube.com
Author
AI Engineer
Date
Why it matters
Giving a coding agent a dependency graph does not guarantee it uses it; tool naming and familiar output formats decide that. The workshop shows evals that check actual tool useA model's ability to call external functions — run code, search the web, edit files — instead of only generating text.Full definition → across repeated runs.
Key takeaways · AI-distilled
Two Rust types both named PostingList show the limit of text search: grep finds both, but a context windowThe maximum amount of text a model can consider at once — its working memory for the current conversation or task.Full definition → graph keeps their identities separate and exposes the distinct code that depends on each, since changing either breaks different things.
For data systems, Michael describes a context layer that computes grain, lineage and relationships from transformation code and checks join assumptions against real data, because valid SQL can still be wrong when a join multiplies rows.
Callender's build over selected Qdrant crates uses Tree-sitter to extract syntax, Rustdoc to resolve symbols and Neo4j to store relationships, so 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 → can ask which symbols depend on a type instead of repeatedly searching files.
Beyond scores, the presenters inspect agent traces, which caught flaws hidden by a favorable evalA repeatable test for AI quality — a set of tasks plus scoring — used the way software teams use test suites, because model output is too variable to judge by eyeballing.Full definition → result.
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.
eval — A repeatable test for AI quality — a set of tasks plus scoring — used the way software teams use test suites, because model output is too variable to judge by eyeballing.
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.
tool use — A model's ability to call external functions — run code, search the web, edit files — instead of only generating text.
videoAI Security Engineer Foundations + Certificate — Micah Silverman, Snyk
videoThe SDLC You Build Yourself — Andy Wong, Andrei Bocan & Shane Wolf, Atlassian
videoAgent Speedrun — Elizabeth Fuentes Leone & Sandhya Subramani, AWS
videoCan Your Agent Hear You Now? Building Live Voice Agents with Gemini — Thor Schaeff
videoAI on Your Lakehouse: Context Comes in Shapes, Not Queries — Zach Blumenfeld, Neo4jAI Engineer
videoHow to Generate Mergeable Code with a Context Engine — Peter Werry, UnblockedAI Engineer
videoYour agents lack context: Here's how to fix "You're absolutely right!" — Brandon Waselnuk, UnblockedAI Engineer