Postgres Full-Text Search in Production: How to Load-Test the Index and Pin Down

<p>Munchable answers one question: does this food fit the gut condition you live with. Until recently there was exactly one way to ask it, which was to hold a phone camera up to a

<p><em>B-tree internals, column order and planner statistics in Postgres 19, checked against MongoDB's ESR rule.</em></p> <p>Most "add an index" advice stops at <code>CREATE

<!-- 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

<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

<p>Your RAG application works perfectly with 1,000 documents. You push it to production, upload 10 million vectors, and suddenly your query latency jumps from 50ms to 5 seconds.

<h2> Search Is Harder Than It Looks </h2> <p><code>LIKE '%query%'</code> works until:</p> <ul> <li>Your table has 100k rows (Seq Scan, slow)</li> <li>Users misspell words (no

<p>We paid for a vector database for a year. It was faster than Postgres at the one thing it does. We deleted it anyway.<br /> </p> <div class="highlight js-code-highlight"> <pre

<p>The chunk was in the index. I checked three times.</p> <p>A support bot kept answering "that isn't covered in the documentation" for a question that was covered, in one