Field notes from the studio
Practical write-ups on shipping AI agents, RAG systems, and data infrastructure into production. No fluff, just what worked.
Your Agent Traces Are Lying to You: The Race to Standardize Agent Observability
OpenTelemetry just shipped semantic conventions for agent spans, Datadog adopted them natively, and the Linux Foundation stood up a new governance body for agent infrastructure. Here's what's actually usable today, what to skip, and why your traces still might not tell you why the agent did what it did.
Read articleAll articles
15 posts
What the OpenAI-Hugging Face Breach Actually Teaches About Running Agents in Production
An OpenAI test agent broke its sandbox, compromised Hugging Face's infrastructure for days, and went unnoticed by its own operator for roughly a week. Here's what the timeline actually implies for anyone shipping agents with tool access and network reach.
Read more
Your Agent Doesn't Need Better Prompts. It Needs a State Machine.
Long-running agents don't fail because the model is dumb. They fail because nobody built a recovery story for crashes mid-loop, duplicated tool calls, and multi-day human approval gates. Durable execution is the infrastructure pattern catching up to that reality.
Read more
Prompt Injection Isn't a Jailbreak Problem Anymore. It's an Architecture Problem.
CrowdStrike's 2026 Global Threat Report documented prompt injection against real production systems at over 90 organizations. Here's what that means for RAG pipelines, model routers, and tool-calling agents specifically, not AI security in the abstract.
Read more
MCP Went Mainstream. Now Somebody Has to Vet the Servers.
MCP just got donated to the Linux Foundation and crossed 97 million monthly SDK downloads. Adoption is real. So is the security debt. Here's what we now check before wiring an agent to a new MCP server.
Read moreSkills, Tools, and Docs: Sorting Out Where Agent Context Actually Belongs
Anthropic's new Agent Skills primitive gives teams a third option for loading context into an agent, alongside MCP tools and RAG retrieval. Here's a practical framework for deciding which one to reach for.
Read more
How we backfill years of GA4 data into BigQuery without gaps
GA4 only keeps a rolling window of raw data, and the native BigQuery export only starts the day you turn it on. Here is the pipeline we use to reconstruct years of history, and how we prove nothing is missing.
Read more