Trust, but Trace

Trust,
but Trace

A journal on what it takes to make ML systems trustworthy in production: traces, fixtures, contracts, replays. Written from the engineering side.

“Most of what an LLM agent does in production has never appeared in any evaluation set.”

Topics

Recurring threads

4 threads that keep coming back. Click one to see every piece tagged with it.

On the desk

Reading list

What’s on the desk this season.

  • 01
    AI Engineering: Building Applications with Foundation Models
    Chip Huyen · 2025
    Book
  • 02
    Latent Space — Artificial Analysis on independent LLM evals
    swyx & Micah Hill-Smith · Jan 2026
    Podcast
  • 03
    Multimodal, Real-Time AI Agent Systems
    Heiko Hotz & Sokratis Kartakis · 2027
    Book
  • 04
    EU AI Act, Article 15: Accuracy, robustness and cybersecurity
    European Commission · 2024
    Reference
  • 05
    Hands-On Large Language Models
    Jay Alammar & Maarten Grootendorst · 2024
    Book
  • 06
    Constitutional Classifiers
    Anthropic · 2025
    Paper
  • 07
    Pragmatic Engineer — Inside engineering at frontier labs
    Gergely Orosz · 2026
    Podcast