#intermediate
Every post tagged "intermediate".
222 posts · showing 181–190
-
Observability for Agents: What to Log Beyond the Trace
A trace explains one run and nothing about your system. The two layers above it — run records and aggregates — plus three agent-specific signals that standard APM will never suggest.
-
The Code-Review Agent That Commented on Everything
Nine comments on a twelve-line PR and the team muted it. Any agent asked to find things will find them — the fix is a hard cap that forces ranking, and pointing it at what humans actually miss.
-
Canary Releases for Prompt Changes
A sentence added to fix one behavior routinely changes three others. Ship prompts like behavior: versioned by rendered hash, split by stable key, and compared on tool distribution — the row nobody collects.
-
Chunking for Agents Is Not Chunking for Search
Search results are read by a person who fills in the gaps. An agent doesn't — it acts on the fragment. That one difference changes how you chunk, how many you return, and what "no results" should mean.
-
Fan-Out, Fan-In: Parallelism Inside a Single Agent
Twelve independent checks shouldn't cost twelve turns, each re-sending the whole context. Three levels of parallelism inside one agent — and the fan-in is where the pattern usually goes wrong.
-
When the Worker Needs to Ask the Supervisor a Question
A blocked worker with no channel back will guess, and the guess is invisible in what it returns. Making "needs input" a result type costs nothing — and the questions it logs are a to-do list for your delegation template.
-
Model Routing: Using the Cheap Model for the Cheap Steps
Classification, extraction, and formatting move to a cheaper tier safely. Tool selection and planning don't — and the failure looks like a tool-description bug, so you'll debug the wrong thing for a week.
-
The Data-Cleaning Agent and the Column It Silently Dropped
It normalized the phone numbers, standardized the country codes, and quietly deleted 412 rows that looked untidy. Any task phrased as "make this better" reads as permission to remove what doesn't fit.
-
Designing a Kill Switch That Works Mid-Run
Killing the process leaves side effects in an unknown state. A stop that works checks between steps — before every model call and before every tool call — and the difference matters most at 2am.
-
Building an Eval Set From Production Traces
Hand-written cases test what you thought of. A user rephrasing the same request three times is a labelled failure sitting in your logs — here's how to turn a week of traces into a suite that catches the real ones.