Query Profiling Goes GA, Weaviate Turns Seven

Hello Weaviate Community! 🤗

Query profiling is now generally available, giving you a per-stage, per-shard view of where a slow query spends its time. We also celebrate seven years of Weaviate with co-founders Bob van Luijt and Etienne Dilocker, and revisit some practical resources for importing data at scale and building with Weaviate.

Latest AI & tech insights

Explore our recent Weaviate content:

Read

• ✍️ Query Profiling: See Where a Slow Query Spends Its Time: Query profiling returns timing data for vector search, keyword scoring, filter evaluation, and object retrieval across every participating shard. Enable it for a single query to diagnose latency without restarting nodes or searching through distributed logs. Read the blog

• 📥 Import & Vectorize Data with Weaviate at Scale: This guide covers production patterns for fast, reliable imports, including server-side batching, error handling, data type decisions, and ingesting media and PDFs without a separate OCR pipeline. Read the blog

Watch

• 🎙 Founding Weaviate with Bob van Luijt and Etienne Dilocker: Weaviate's co-founders answer community questions and revisit seven years of building the company, including the early NodeJS prototype, the move to Go, and the decision to adopt HNSW. They also discuss agentic coding, world models, open source models, and why human taste matters as AI-generated work becomes more common. Watch the full podcast

• 🎥 How to build agentic RAG systems that actually work: Victoria breaks down how context shapes every stage of an agentic RAG pipeline, from chunking and short/long term memory to tool use and the orchestration layer that ties everything together. Watch the full video

Product highlights

Check out our latest product updates.

• Query profiling (GA): Add an opt-in flag to a single query and receive its timing breakdown inline, collected across every shard and node involved. It is available through the official Python, JavaScript/TypeScript, Java v6, and C# clients, as well as gRPC. See how it works

• Managed Weaviate on DigitalOcean (Public Preview): Spin up a Weaviate cluster without operating it yourself on Kubernetes or VMs. DigitalOcean manages the infrastructure, while remaining fully compatible with the open-source Weaviate engine and existing client libraries. Read more

• Introducing the Weaviate Playground: Explore all the interactive demos built with Weaviate, including the Query Agent, multimodal search, AI-powered recommendations, and Engram. Explore the playground

Company updates

Join our team - We're hiring across various teams! ➡️ Check out our career page for opportunities in engineering, research, and more ✨

Hungry for more?

Have a question or want to connect? Join our Weaviate Forum to engage in community conversations.

Follow our Blog, LinkedIn, Twitter, and YouTube for more content updates.

See you in two weeks,

Prajjwal