Memory & Context
23 posts · showing 11–20
-
Was It Retrieval or Reasoning? Diagnosing a Wrong Answer
A day spent rewriting the prompt for a wrong policy answer, when the policy document was never retrieved. Three checks in the right order, and one log field that separates "ignored it" from "never saw it."
-
Budgeting the Context Window: Dividing a Fixed Space
The window wasn't full and the constraints still vanished — because six claimants compete for one space and nobody allocated it. Making history the residual rather than a claimant is what makes the scheme work.
-
Context Engineering Is the Job Now
Not prompt writing — that's the visible part and the small one. The work is deciding what reaches the model, in what form, at what moment, and it's where most agent failures start.
-
Longer Context Windows Won't Fix Long-Horizon Tasks
The premise is right — things fall out of context — and the conclusion doesn't follow. Most long-run failures aren't about storage, and a bigger window removes the discipline that was helping.
-
One Query Is Rarely Enough: Iterative Retrieval for Agents
The search wasn't broken — it answered the question as typed, which isn't the question that was asked. Making retrieval a tool the agent calls repeatedly needs one signal most retrievers refuse to give.
-
When Memory Makes an Agent Worse
Every retrieved memory is content injected into the context, relevant or not. Three ways that degrades the agent — including the one where it argues with a user about their own stated preference.
-
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.
-
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.
-
Run, Session, or Global? Scoping Agent Memory Correctly
Too-small a scope produces a visible annoyance; too-large produces a silent wrong answer. One question assigns any fact correctly, and one rule keeps multi-tenant reads from becoming an incident.
-
Where You Put Things in the Context Changes the Answer
Same instructions, same documents, different results — because the constraint sat in the middle of a long array. Two ordering rules, one of which also decides whether your prompt cache ever hits.