# AIE CODE: OpenAI, ArtCraft, Cognition, Exa, Anthropic, Amp, et al

Source: [AI Engineer](https://www.youtube.com/watch?v=y9JtCCZP2MQ)  
Feed7 permalink: https://feed7.dev/p/aie-code-openai-artcraft-cognition-exa-anthropic-amp-et-al-0gc85v9  
Published: 2026-11-13T02:12:27.000Z  
Trust: Source Linked (source_linked)

## Why Included

One conference discussion recommends mining agent traces for reusable skills and evals before fine-tuning, with nearly 80,000 team sessions as source material.

## Source Summary

A speaker described using agent-session traces to create reusable skills and then build evals from them. Their team had **almost 80,000 sessions** available as source material.

## Practical Implication

For builders, the practical sequence is to inspect traces, write a skill for a recurring task, and evaluate it before considering specialized fine-tuning. Traces can also reveal when agents recreate libraries or produce work that needs stronger review.

## Agent-Ready Context

A speaker described using agent-session traces to create reusable skills and then build evals from them. Their team had **almost 80,000 sessions** available as source material.

For builders, the practical sequence is to inspect traces, write a skill for a recurring task, and evaluate it before considering specialized fine-tuning. Traces can also reveal when agents recreate libraries or produce work that needs stronger review.

The transcript combines several sessions and is fragmentary, so it does not provide measured results for this workflow. The session count is one speaker’s team context, not a general benchmark.

## Connected Context

Feed7 judgment across 1089 accumulated Signals:

This gives a practical source for reusable skills and evaluations: inspect accumulated session traces, turn recurring work into skills, and derive evals from those traces. The nearly 80,000 sessions describe one team’s material, not evidence that the approach improves results generally.

- [How We Solved Agent Building — Andrew Qu, Vercel](https://feed7.dev/p/how-we-solved-agent-building-andrew-qu-vercel-1qhqr4p) — Vercel’s account supplies internal eval evidence for reusable filesystem context, while this talk suggests traces as material for building and evaluating such workflows.
- [fighting slop with slop — Vaibhav Gupta, Boundary](https://feed7.dev/p/fighting-slop-with-slop-vaibhav-gupta-boundary-0w6qoke) — Boundary also uses transcripts and execution traces to improve agent workflows, reinforcing trace inspection while retaining human triage.
- [tech-leads-club/agent-skills](https://feed7.dev/p/agent-skills-1rcu2i7) — Turning recurring work into reusable skills raises a distribution and trust concern that this registry addresses with scanning, lockfiles, and audit trails.

## Context Map

- Layer: agent
- Domains: coding
- Topics: skills, agent-evals, harness-engineering

## Uncertainty

- The transcript combines several sessions and is fragmentary, so it does not provide measured results for this workflow. The session count is one speaker’s team context, not a general benchmark.

## Agent Instruction

Use this item as source-backed context. Do not invent claims beyond the linked source. If this item conflicts with another source, call out the conflict.
