Similarity isn't relevance: the hard part of semantic search

<p>Most semantic search systems don’t fail because of embeddings. They fail because of how the system is designed around them.</p> <p>When I started building a semantic search API

<!-- SC_OFF --><div class="md"><p>We spent the last year building a code indexing system and tried pretty much every retrieval approach along the way. Sharing what actually worked

<!-- SC_OFF --><div class="md"><p><strong>Cracking ML System Design Interviews — Design a Search and Ranking System</strong></p> <p>I recently wrote Part 2 of my ML System Design

<p>In the rapidly evolving landscape of Generative AI, the transition from experimental Proof of Concepts (POCs) to production-grade applications is the most significant hurdle

<p>You spent months tuning your <code>&lt;title&gt;</code> tags, chasing backlinks, submitting sitemaps to Google Search Console. Your rankings are solid. Then you ask ChatGPT

<p>I asked five AI search engines the same question: "What's the best CI/CD tool for a small team?" Google AI Overviews recommended GitHub Actions. ChatGPT went with GitLab CI.

<p>Every RAG prototype wants a vector database before it wants a data model.</p> <p>That is usually backwards.</p> <p>The first hard question is not which ANN index has the nicest

<p>Over the last month, I’ve been working on a personal project: building a search engine from scratch.</p> <p>This started from a simple curiosity — I’ve always wanted to