Writing
Notes on memory, mid-build.
Honest field notes on how AI memory actually behaves once you take it beyond a single chat — what compounds, what decays, and where the public debate lines up with what we've shipped.
-
13 November 2026
Memoria on three public memory benchmarks: what we measured and what we found
We ran Memoria through OmniMemEval, the open harness that tests memory systems with the same models. It placed fifth on LoCoMo, third on LongMemEval-S and second on BEAM-100K. Its context is small on one benchmark and large on another. Its knowledge graph lowered the LoCoMo score and made no difference on LongMemEval-S.
Read the post
-
8 July 2026
Availability was the loud story. Sovereignty was the quiet one.
In June a frontier model was altered, withdrawn, then restored a few weeks later. The lesson everyone reached for was about models. For an Australian regulated buyer, the one that matters is about substrate.
Read the post
-
9 May 2026
Pinecone is moving past RAG. The hybrid is what's next.
Pinecone shipped Nexus, a compile-then-serve substrate that looks a lot like the architecture Memoria's been building. Same shape; different posture on what the substrate is for. Where we agree, and where we diverge.
Read the post
-
29 April 2026
Du's agent memory taxonomy and the org-wide memory gap
Pengfei Du's new survey gives agent memory a proper academic frame. Memoria's hybrid memory infrastructure maps onto it cleanly — and the org-wide gap the taxonomy surfaces is the part most current tooling doesn't reach.
Read the post
-
27 April 2026
The hybrid memory architecture (and why it took a public debate to name it)
The hybrid memory architecture Karpathy and OpenBrain are converging on — and the auditable retention layer the debate hasn't reached yet.
Read the post