#agent-design
Every post tagged "agent-design".
112 posts · showing 71–80
-
Request, Worker, or Durable Execution? Hosting Models for Agents
Where a run lives decides how long it can be, whether it survives a deploy, and whether it can wait for a person. Pick by the longest wait it can hit — and keep state out of process memory so the answer can change.
-
Designing Tools for Agents vs Designing APIs for Developers
One endpoint, one tool is the obvious mapping and the wrong one. APIs assume a consumer who reads docs once and writes a loop — an agent re-decides everything from a paragraph, every call.
-
Getting an Agent to Say How Sure It Is
Ask a model how confident it is and you get a plausible number that doesn't move with difficulty. Five signals that actually correlate with being right — and the calibration check that tells you if yours is noise.
-
Few-Shot Examples: When They Help an Agent and When They Hurt
Two or three similar examples is the worst case — enough to teach a template, not enough to convey the range. And the example nobody includes is the one where the right answer is to decline.
-
The Documentation Agent That Actually Kept Docs in Sync
Ask an agent "do the docs need updating?" on every PR and you get noise reviewers close unread. Trigger on contract changes, forbid new prose, cite the source line — and run the corpus-wide contradiction check first.
-
Delegating to a Human as If They Were an Agent
Same interface, same contract, same result shape — a person is a worker with hours of latency and much better context. Which makes suspendable runs the prerequisite, not a nice-to-have.
-
The File System as a Tool: Underrated Agent Infrastructure
Four hundred thousand rows won't fit in a context window; a path and a preview cost nothing. A scoped directory solves large results, durable state, and reviewable artifacts at once — if you resolve paths correctly.
-
State Machines for Agents in Regulated Flows
When doing steps out of order is a compliance failure regardless of outcome, a model-decided control flow is the wrong shape. Put the ordering in code, the conversation in the agent, and scope the tools per state.
-
Tool Calls or Structured Output? Two Mechanisms, Two Jobs
Both emit schema-conforming JSON, which makes them look interchangeable. One expects a result back and one ends the turn — and that single difference decides every other question.
-
Give the Agent a Scratchpad: External Working Memory
Three small tools stop the context from being the only place findings can live. Compaction stops being lossy, long runs stop growing, and you get a readable record instead of a forty-turn transcript.