Christophe Ponsart

Founder Former founder Technology Early Adopter

[Christophe Ponsart @cponsart]

@gheeunit @johnternus Yeah (good eye). It was great seeing him making the rounds

[Christophe Ponsart @cponsart]

@adcock_brett Yep, Sold!

[Christophe Ponsart @cponsart]

@Jason We just did Sumo show today and did Samurai classes yesterday. It was great for our kids 18yrs through 5yrs old!

Let me know if you’d like the name of the spots we hit? They were very highly regarded

[Christophe Ponsart @cponsart]

@DataChaz @3blue1brown @DataChaz Completely agree. I built ContextFit based on this idea (tokens are better for memory) and the memory system is hitting top marks on storage, performance and retrieval. 99.0% longmemeval. Check out the approach at https://t.co/xDf0FhRo4U

[Christophe Ponsart @cponsart]

@steipete @RoccoScalps @openclaw Great! Love the openclaw + telegram integration!

[Christophe Ponsart @cponsart]

@puresight Not yet 😁

[Christophe Ponsart @cponsart]

AI memory shouldn’t require a vector DB by default.

ContextFit keeps memory token-native: token chunks, metadata, source grounding, transparent retrieval.

Vectors/BM25 can still be fused in as optional signals when they help.

Latest pass: 96.6% R@5, 98.7% R@10.

AI memory should be token-native.

[Christophe Ponsart @cponsart]

Recency is a signal, but not the primary policy.

Fusion should be type-aware: a preference update, decision record, operating policy, and observation shouldn’t all compete on the same axis.

A newer preference may override an older preference, but it shouldn’t silently override a decision record. It should create a conflict edge: “this preference contradicts this decision.” Then resolution depends on type, authority, scope, and approval state.

So less “latest wins,” more “typed memory with provenance + explicit conflict handling.

[Christophe Ponsart @cponsart]

ContextFit just hit 99.0% retrieval on LongMemEval-S n=500.

The unlock: memory atoms + fusion. Instead of treating memory as one flat vector search problem, we route preferences, decisions, temporal updates, and open loops differently.

Agent memory needs structure, not just more context.

[Christophe Ponsart @cponsart]

Exactly. The categories are routing hints, not hard walls. A decision can also be a preference signal, and a temporal update can create an open loop.

The interesting part is reconciliation: retrieve/fuse across typed atoms, then let source/provenance + recency decide what survives. Otherwise you just move the brittleness from vectors into labels.

[Christophe Ponsart @cponsart]

@LearnWithBrij @LearnWithBrij Agree with all of these. Can I add one more consideration? I’m finding strong outcomes with a new approach which uses token-based memory retrieval instead of just vector based retrieval. Would love your feedback - https://t.co/28RQwfvTLZ

[Christophe Ponsart @cponsart]

@henrytdowling Here are our latest benchmark details: https://t.co/1BrxEa9tpj

[Christophe Ponsart @cponsart]

The 98.7% is a token-based and vector fusion approach for even higher quality outputs

[Christophe Ponsart @cponsart]

@henrytdowling Longmemeval. I’ll be updating GitHub with the benchmark details

[Christophe Ponsart @cponsart]

Exactly. The interesting part is deciding what should be cached, what should be retrieved, and what should remain directly addressable.

I’m increasingly convinced agent memory should stay token-native at the core: precise, source-grounded, and cheap to search — with RAG/CAG/fusion layered on top when needed.

That’s the direction we’re taking with ContextFit.

[Christophe Ponsart @cponsart]

@garrytan @garrytan is there an available public dataset I can run my token-based memory system against to compare stats? I built a memory retrieval system that is 50x faster than vector embedding search and requires no database, no vector embeddings.

[Christophe Ponsart @cponsart]

Completely agree with this. Vector response time and interpretability are not the solution for ai agents.

grep is the right direction being file system native but provides no assistance with discoverability outside of direct keyword search.

I believe the solution is token-native, file system native (requiring no database), and the results of my test so far are blowing my mind. Let me know what you think of this approach?

https://t.co/1wDi7kQ3lr

[Christophe Ponsart @cponsart]

I made an interesting discovery that could potentially have a significant impact to the AI industry! I'm open sourcing the concept and code on github via MIT license in an effort to validate if this approach is meaningful or if I've lost my marbles.

Watch the video and tell me what you think? Once you hear it, it's hard to un-hear... "Should agentic memory live in tokens and not vectors!"

If you believe this idea, help me spread the word by replying here.

Also, if you're a developer, please download/star the repo. Would love your input and feedback.

This is my first open source project (so yes, I have no idea what I'm doing :)

ContextFit retrieves the correct answer from your existing knowledge base and conversations without embedding APIs, vector databases and at 50x the speed, significantly improving agent workflows by reducing both time and cost.

[Christophe Ponsart @cponsart]

Got my first GitHub star today.

Tiny number. Weirdly meaningful.

I’ve spent my whole career around consulting companies, where most of the best work happens behind client walls and never becomes open source.

So seeing someone outside the room find a project useful enough to star it feels different.

Small milestone, big feeling.

[Christophe Ponsart @cponsart]

I’m starting to write publicly about ContextFit, an open-source project I’ve been building around token-native memory retrieval for AI agents.

First piece: why I think AI memory needs to move beyond “embed everything and hope the nearest chunks are enough.”

AI memory should be fast, local, auditable, and built for the way models actually consume information: tokens.

[Christophe Ponsart @cponsart]

https://t.co/1iXolJjpK4