» Tag
embeddings
11 postsAI Agent Memory Is Just a Smarter Filing Cabinet
Today's AI agent 'memory' is retrieval in disguise—it can't tell if past answers were right. Real memory needs judgment that learns from outcomes.
Vector Database Internals: From Brute-Force Search to IVF Indexing
A technical look at vector DB schema design, why brute-force search fails at scale, and how IVF indexing narrows search to relevant clusters.
Sentry's New AI Model Cuts Issue Grouping Errors in Half
Sentry upgraded its AI-powered issue grouping model, cutting duplicate issue creation by 20% and halving the rate of incorrect merges in production.
CommitBrief — AI code reviews, right in your terminal
A provider-agnostic, local-first CLI that reviews your staged changes, a historic range, or a whole GitHub pull request. Zero telemetry, no server. Free and open source.
commitbrief.comWarner Chappell Cuts Cloud Costs by 50% With AI Embeddings
Warner Chappell replaced brittle SQL matching with in-memory AI embeddings, cutting cloud costs 50% and boosting royalty-matching output by 65%.
rememori: A Zero-Dependency TypeScript Memory Engine for AI Agents
rememori is a zero-dependency TypeScript library giving AI agents persistent memory by combining cosine similarity, entity graphs and time decay.
A Fake Python Memory Leak: ru_maxrss vs VmRSS Confusion
A RAG pipeline seemed to leak memory as RSS kept climbing, but the real bug was measuring ru_maxrss (a peak value) instead of live VmRSS. Fix, baseline check, and batching included.
Better Embeddings Won’t Fix Missing Provenance in RAG
Explore how missing provenance can lead to inaccuracies in retrieval systems, despite better embeddings.
Production RAG at Scale: Lessons from Processing 10,000+ Listings Daily
Insights on RAG pipeline stability, chunking strategies, and cost control from processing job listings.
SigLIP 2 Text Embedding Server on CPU with Rust and ONNX
The CPU-only SigLIP 2 text embedding server, built with Rust and ONNX, is ideal for live queries when GPU resources are limited.
Building a Multimodal Vector Search Platform for Product Catalogs
How a recommendation team built a multimodal vector search platform with Qdrant, tuned HNSW settings, and optimized GPU usage for embedding millions of products.