Agent Infrastructure · Startup idea
Agent memory stores: the database the LLM remembers
Vector databases solved retrieval. Nobody solved memory, the layer above retrieval that knows what the agent already learned, forgot, and should re-check. That's the next category.
Why now
Every agent in production runs into the same problem: the context window fills up, the agent loses track, the user repeats themselves. The fix is a separate memory tier, write-on-significant-events, decay-by-recency, query-on-need.
The idea you could build today
A managed service that exposes three tools: `memory.write(event, importance)`, `memory.recall(query)`, `memory.forget(predicate)`. Backed by a graph + embedding store. MCP-first so any agent host can plug in.
Build stack
- ·Postgres + pgvector OR Turbopuffer
- ·Embedding generation via OpenAI or Voyage
- ·MCP server for the read/write surface
- ·Inngest for the background re-indexing
The three repos already trying
Pulled live from our current-period signal index.
+67%
14-day velocity Δ
5 contributors
Real-time data integration platform connecting databases, services, and message queues through a unified streaming pipeline.
Engineering hiring burst
+421%
14-day velocity Δ
23 contributors
Bruin is an end-to-end data platform with built-in data quality, observability, and governance.
Engineering hiring burst
+19%
14-day velocity Δ
13 contributors
Matched against the current-period startup signal panel (ai-ml, data-infrastructure). Rankings shift weekly as the underlying GitHub activity moves. Read the methodology.
The seed-round pattern hiding in the trendline
Memory-layer OSS projects (Mem0-class, MemGPT-class) with sudden contributor surges around the "importance scoring" or "decay" modules are the seed-round leads.
Frequently asked
Isn't this just RAG with extra steps?+
RAG is read-only. Memory is read-write with importance scoring and decay. Different write path, different query semantics, different product.
Use the signal, not just the idea
Watch this idea live, every week.
The repos above re-rank automatically as commit velocity, contributor growth, and new-repo creation move. Want the data feed for this idea wired into your own stack? The MCP server exposes every signal as a tool any agent host can query.
Related ideas
Updated 2026-08-25. The framing is editorial; the “three repos already trying” slot is generated from the live signal panel. Anonymity rule: we name public GitHub orgs, never individual founders or stealth teams.