Topic hub · 6 pieces

Agent memory

Agent memory names at least four technical products: context graphs, vector retrieval, continual learning and persistent session state. The first memory piece here argued that a vector database earns its place only past roughly 50,000 episodic items, and Letta made a git-backed markdown memory, MemFS, its default in February 2026.

The money arrived in June. Jedify raised a $24 million Series A and Relai announced $6.9 million on June 10, and Engram left stealth on June 23 with $98 million. In the same quarter OpenAI and Anthropic shipped memory as a built-in feature of their own products.

Timeline

  1. · Agents

    Memory engines for long-running agents

    Every agent-memory tutorial opens with a vector DB, but that is a product opinion rather than a default; mature agent runtimes start elsewhere, and Qdrant or sqlite-vec earns its keep only past roughly 50K episodic items.

  2. · Tools

    gbrain: the Postgres-native knowledge graph Garry Tan runs his AI agents on

    gbrain, the persistent memory layer behind Garry Tan's own AI agents, runs 66 autonomous cron jobs over 146,646 pages and 24,585 people, making it the rare self-hosted alternative to vector-DB wrappers in 2026.

  3. · Agents

    The agent-memory funding wave hides four different bets

    On one June Tuesday, money poured into startups all selling AI-agent 'memory,' a word that covers at least four different technical products, while the frontier labs absorb the easiest of them as a free feature.

  4. · Agents

    Supermemory open-sourced its memory engine. The moat moved somewhere else

    On June 10 Supermemory shipped its graph engine as a local binary anyone can run for free, and reading the repo, the docs, and the benchmark fight shows the giveaway is a distribution play, not a surrender.

  5. · Tools

    Mastra in production: the advanced patterns worth shipping

    An operator playbook for Mastra across six TypeScript codebases and the full 1.x line, covering memory's four layers, supervisors, MCP on both ends, RAG and evals, auth, observability, the 1.x migrations and the production bugs that taught the most.

  6. · Business

    Engram raised $98M to give enterprise AI a memory. Where's the moat?

    Engram left stealth with $98M at a reported $600M valuation and 13 employees to build a learned memory layer that studies an organization in advance instead of re-reading its documents, a real technical bet with a harder moat question.

In the data

Rows from the CC BY 4.0 datasets that these pieces report. Each links to its row, primary source and method.

Funding and acquisitions · 3

Key primary sources

More on this subject

Tag: agent-memory Tag: memory Department: Agents All topic hubs