Topic, page 2
Page 2 of articles on AI.
Examples are the fastest way to fix a format problem and a slow way to fix a reasoning one. Adding more has a ceiling that arrives earlier than expected.
Teams reach for fine-tuning to make a model know their data. It is much better at teaching behaviour, and facts have a cheaper home.
Prompts, traces, embeddings and evaluation sets all quietly become copies of user data, and most of them were never on anybody's data map.
Agents rarely crash. They loop, they stop early, and they confidently do the wrong thing, which is why ordinary error handling never catches any of it.
Approximate search is approximate on purpose. Knowing which results you are giving up explains retrieval failures that look like model problems.
People are the only judges who catch what nobody thought to check for. Spending that attention on random samples wastes most of it.
Vector search handles paraphrase well and exact terms badly. Keyword search is the reverse, and users send both kinds of query in the same session.
Most RAG systems are built against a snapshot and then quietly decay. Indexing is a pipeline with a freshness target, not a one-off import script.
Grading output with another model scales in a way humans cannot. It also has biases of its own, and a judge nobody validated is just a number.
Conversation history, retrieved knowledge and learned state all get called memory. They have different lifetimes, and conflating them is why agents drift.
Real queries are never purely semantic. They are scoped to a tenant, a status and a date, and where that filter runs changes the results you get.
The capability gap narrows every few months. What does not narrow is the difference in who controls deprecation, data residency and your cost curve.
Whether it needs designing from scratch or rescuing from its own success, tell me what you’re building and I’ll tell you how I’d architect it.
Open to remote and hybrid work worldwide