SeriesAI FoundationsPart 4
Field Notes: Six 2025 Ideas That Changed How I Build RAG & Agents
A story-driven tour of new papers and benchmarks—plus what actually moved the needle in my pipelines, with links and practical takeaways.
Two bugs broke my RAG system last month. The first was self‑inflicted—I nudged top‑k from 8 to 16 “for coverage” and watched answers get longer and less faithful. The second came from a knowledge base that contradicted itself; my agent chee...