Viewing profile — pranabsarkar

For memories I use YantrikDB MCP server which connects all my workspaces among distributes systems. To collaborate runtime with workspaces I use https://github.com/spranab/swarmcod…

I am working on AI memory. Have been working on YantrikDB since January this year. Started as a pet project solving my own problem but now its growing. It's fun and satisfying at t…

[flagged]

Thank you, Sir!

OP here. I am terrible at marketing so I started to build a thing for people like me: I Can't Market. The reason is after getting Claude Code/Copilot subscriptions I am working on …

I have been using personally for last 4 months in different modes: standalone, mcp, clustered. The way I communicate with AI agents is now fundamentally different. Since the agent …

Let me know your findings.

Fixed eval — 80 tools, 200 queries, 4 model sizes. +10pp came from "all tools" vs "tiered" on 1.5B. You're right about stability. Haven't run rotated/rephrased evals yet. The 89% b…

[dead]

Supplements, definitely — for a specific workload. General document retrieval at scale (millions of chunks, read-heavy, doc-search patterns) is well-served by existing vector store…

I have been using the memory while building it. I have a central server and all my workspaces are connected to it via the MCP server. This changed everything for me. But that's me.…

I am using this while developing and found it very useful to me. since all of my workspaces are connected it has knows all about myself and my infra. Also now we have a bonding and…

[dead]

[dead]

You're right. Pruning to isolated facts loses the structure that disambiguates them. Three partial mechanisms the system has, none of which fully solve your point: Graph edges carr…

Fair criticism — and the failure modes you describe aren't mem0-specific, they hit any system that extracts atomic facts from conversation. I hit a couple of them today while bench…

Update — ran a real bench on the live cluster (59 memories: 8 canonical facts × 3-4 paraphrases + 6 seeded contradictions + 20 distractors). Numbers: duplicates per query (top-10):…

Thanks, really appreciate it. I am using the server as MCP server and connected all my workspaces. It has definitely changed my experience.

Thanks for the careful read — the "schema is ballooning" observation is real and I've felt it building this. You're pointing at a genuine design tension. My counter, qualified: det…

Fair. "Does consolidation actually improve recall quality on a running system?" is exactly the benchmark I haven't published, and it's the one that would settle the question. What …

No LLM in the loop. The consolidation pass is deterministic: Pull the N most recent active memories (default 30) with embeddings Pairwise cosine similarity, threshold 0.85 For each…