There's a class of search problems where: The total number of records is small ~1 million The index is built up front and doesn't change frequently Don't get hoodwinked into buying a vector database. Just brute-force search your embeddings. A naive Numpy search can do exhaustive vector search at 170QPS. You can replicate and copy that data to scale even further. More in my blog article: https://softwaredoug.com/blog/2026/07/29/just-brute-force-embeddings Wectors Week + Context Camp A reminder...
18 days ago • 1 min read
Almost all my clients face this issue. They want to build a RAG system. They have: Millions of PDFs That turn into Billions of chunks + embeddings Turns into become petabytes of floats That's the recipe of nearly every speculative RAG project I've worked on. A mature search team stands up a vector database and can't tame the sheer scale of data they're facing. That drags project timelines out months while costs skyrocket. It's important to have tools to manage the scale. One often overlooked...
21 days ago • 1 min read
An agent with grep does quite well at search. It’s shocking to search technologists. We want to cocoon the problem in technology. We care about algorithms, knowledge graphs, ranking: ever-and-ever smarter retrieval. It turns out, though, you can play a bit of a trick. If you convince everyone to optimize content for your search engine, you’ll have built the best search engine. There’s Sutton’s bitter lesson about unleashing raw compute on a problem. But there’s a different bitter lesson in...
about 2 months ago • 1 min read
Agentic search gets interesting when agents do not know how to find the right answer. Oh, the agent might think it knows. It might confidently BS us. But the agent’s poor domain intuition steers itself astray. Agents make false assumptions about what our users think is relevant. Our fashionista users think “red shoes” should return high-heels. When I worked at one company ABE wasn’t a president, it was an A/B testing tool. Agents need context to know these things - and context engineering...
2 months ago • 1 min read
Upcoming events in the next week or so Show us your skills w/ Hugo Bowne-Anderson Thursday May 28th - https://luma.com/ltpzpqgw Pray to the demo gods! I'll be joining Hugo Bowne-Anderson's "Show us your skills" event on Luma - highlighting using a coding agent to optimize search rankers.. Come hang out if you want to see how others in the industry leverage agentic AI to build in their domain. User search trends in 2026 Monday June 1 -...
3 months ago • 1 min read
At Haystack I spoke about autoresearch: Code generation to optimize search rankers. Can we use it to improve on BM25? This article represents my lab notes. My agent starts with a BM25 implementation, proposes changes, and accepts those that improve NDCG. We’ll zero-in on passage retrieval dataset MSMarco. I won’t claim I’ve found a “better BM25” but I’ve iterated towards a decent tuning regime. All while learning valuable lessons about how validation data can leak. Let’s walk through what...
3 months ago • 1 min read
Search events this week! Bag of Documents Model w/ Daniel Tunkelang Tuesday, 1PM ET - https://maven.com/p/7270ba/a-bag-of-documents-model-for-query-understanding-retrieval Tomorrow, Trey Grainger and I will host Daniel Tunkelang as he introduces his "Bag of Documents" technique for vector search. Bag of Documents is a form of pseudo-relevance feedback using document embeddings. Search week in Berlin June 7-12 2026 - https://berlinsearchweek.com/index.html A reminder in June: Berlin Buzzwords...
3 months ago • 1 min read
It's the final week before my Cheat at Search with Agents course. But we've got a fun tailgaite party before the big event 😀. Agent harnesses are dead - long live harnesses Today, 7PM ET - https://maven.com/p/6dc9ef/building-effective-agent-harnesses Search backend design means agent design. Hugo Bowne Anderson lives by a KISS ethos - Keep it Simple Smartypants. He talks real-life production LLM + harness design. You don't need to chase the latest OpenClaw -> Hermes -> Claude Code design...
3 months ago • 1 min read
Give an agent a set of search tools, it finds relevant products and improves result ranking. So should we throw away our traditional search stack and just let an agent drive some retrievers? Will the future of search APIs just be an agent, not query understanding or reranking? Here's the rub - finding things with agent's help differs from helping agents find information. In one case, the agent helps us. In the other, we must help the agent find what it doesn't know. This last case can't work...
4 months ago • 1 min read