LoginSign Up
← Back to all sparks
W

Weaviate

DEVOPS
Velocity7.5

Open-source vector database for AI applications and semantic search

weaviate.io ↗

Weaviate is rebuilding around agent memory and MCP, not just vector storage.

agent-memoryengrammcp-servervector-databaseragmanaged-cloud
◆Current state
Weaviate's recent feed is anchored by two strategic releases: the 1.37 release with a built-in MCP Server, Diversity Search, and Query Profiling, and Engram — a managed memory service for agents. Surrounding work makes the AI-native database real on more clouds (Shared Cloud GA on AWS US-East and Europe) and surfaces (C# managed client, hybrid-search tokenization improvements). Engineering blogs lean into RAG quality and multimodal embeddings.
◆Where it's heading
The product is rotating from 'vector database' positioning toward 'memory and retrieval substrate for AI agents.' The combination of MCP server in core, Engram as a managed offering, and dogfooding inside Claude Code suggests agent memory is the next category Weaviate intends to own — distinct from raw vector storage, where Pinecone and Pgvector continue to crowd the market.
◆Prediction
Expect Engram to expand integrations beyond Claude Code (Cursor, Cline, custom agent frameworks) and a clearer pricing surface for memory-as-a-service. The MCP server in 1.37 should evolve from preview to GA with curated tool catalogs.

◆Recent moves

  1. 7d ago

    Text Analysis for Hybrid Search: Tokenization, Stopwords & Accent Folding

    Engineering deep dive on multilingual tokenization, accent folding, and the /v1/tokenize endpoint. Useful for developers tuning hybrid search but not a release.

    View source ↗
  2. 15d ago

    Your LLM Is Only as Good as What It Retrieves

    Researcher-perspective essay on retrieval quality in RAG. Reinforces Weaviate's positioning as the retrieval layer for serious AI builders but no product update attached.

    View source ↗
  3. 28d ago

    Weaviate 1.37 Release

    ⚡ SPARK

    Weaviate 1.37 ships a built-in MCP server, Extensible Tokenizers, Diversity Search, and Query Profiling — the most directional release of the cycle, repositioning the database as an agent-native substrate rather than a raw vector store.

    View source ↗
  4. 1mo ago

    Engram: Memory by Weaviate

    ⚡ SPARK

    Engram launches as a managed memory service for agents, designed to be simple to start but adaptable to varied use cases. A new product line distinct from the core database, aimed squarely at the AI agent memory category.

    View source ↗
  5. 1mo ago

    Weaviate Shared Cloud now generally available on AWS

    Weaviate Shared Cloud reaches GA on AWS in US-East and Europe. Makes the managed offering procurable in the regions enterprise customers care most about.

    View source ↗
  6. 1mo ago

    Oh Memories, Where'd You Go

    Two-week dogfooding report on Engram inside Claude Code sessions — useful both as case study and as a frank account of where memory products break down with coding assistants. Reinforces Engram positioning without being a new release.

    View source ↗