Multi-agent systems don't fail. They drift.
A2A solves how agents talk to each other. It doesn't solve how they stay pointed at the same outcome. Each agent can do its job correctly and the system still produces the wrong result.
writing
Things worth writing down.
A2A solves how agents talk to each other. It doesn't solve how they stay pointed at the same outcome. Each agent can do its job correctly and the system still produces the wrong result.
Model Context Protocol gets described as a way to give LLMs access to tools. That framing buries the harder question: how does a system decide what a model is allowed to touch?
Reasoning chains aren't just for accuracy. In an enterprise system they're how you earn trust, debug failures, and catch a model doing something it was never asked to do.
Generation quality is the part everyone demos. Retrieval quality is the part that decides whether the demo was a lie. Notes on auto prompt-tuning for document selection.
The most useful mental model I've found for securing enterprise LLM systems: know exactly which bytes of your prompt the user is allowed to influence, and treat the rest as untouchable.
What I learned building SIGN-LLM: why sign-language generation is really a data problem, and why separating 'how to represent motion' from 'how to produce it' is the trick that makes it work.
Search quality rarely fails loudly. It drifts. Notes on benchmarking, fluctuation analysis, and root-cause work on a recommendation and search pipeline.
A search system you can't measure is a search system you can't improve. Notes on building human-in-the-loop feedback for an enterprise RAG pipeline.
Doing OCR and entity extraction on health records inside a Trusted Research Environment, where the data can't leave and most of your usual tools can't come in.
No posts match that search.