CortexDB.ai by creator of Cassandra@cortexdbai

AI loses context. CortexDB fixes that. Durable memory for agents, copilots & assistants. SDKs, APIs & MCP-ready. By @pmalik

cortexdb.ai
Joined March 2026
20 Photos and videos
Article

The Memory Wars Are Coming

Memory is sticky. Leaving the memory provider means lobotomizing the agent. Inference is commoditizing. Today's frontier model is tomorrow's free tier, and compute providers are locked in a structural

11
8
83
2,238,935
RAG retrieves history. Memory maintains continuity.
Made with AI
1
2
4
405
cortexDB stores memory as an append-only event log, so a wrong answer comes with a source, not a shrug.
Made with AI
2
3
140
coding agents forget everything between sessions. give them one shared memory instead. local and inspectable, built for agent fleets. cortexdb.ai/download
15
13
122
1,696,755
CortexDB.ai by creator of Cassandra retweeted
OpenRouter
@OpenRouter
Jun 13
Introducing the Fusion API, the smartest compound model in the market. Fusion achieves Fable-level intelligence at half the price. How it works 👇
720
1,764
14,985
6,435,189
5/6 this is becoming procurement, not a feature. 88% of orgs hit an ai-agent incident last year, only 21% could see what their agents were doing. eu ai act article 12 makes event logging mandatory for high-risk systems in aug 2026.
1
1
100
6/6 trust used to be asserted: a benchmark, a brand, a promise. traceable memory makes it queryable. "show me the events under this conclusion." cortexdb is that bet: every input stored as an immutable event, never rewritten, everything else derived on top.
67
4/6 the honest limit: reading the memory store is not reading the model. we can see what the agent knew. we can't fully see why it acted. a wrong belief is auditable. a wrong action built on correct beliefs is not, yet.
1
56
2/6 every belief humans hold comes with a story assembled after the fact. the chain that actually produced it was never stored anywhere we could read. agent memory is the first tech that stores it.
1
65
3/6 so the first cognitive system whose reasoning can actually be read won't be human. it's readable precisely because it lacks the interpreter that fills our gaps silently. ask why a belief is held and either the chain is there or it isn't. the absence is visible.
1
52
TLDR: 1/6 ask anyone why they did something and you get a fluent, confident answer. it can be entirely made up. the split-brain work showed it at the wiring level: the brain invents a reason, reports it as the cause, and never knows it's inventing.
CortexDB.ai by creator of Cassandra@cortexdbai
Jun 8
Article

The Legible Mind

Minds confabulate. Every one of them, constantly, and none of them can tell. Ask a person why they did something and the answer arrives fluent and confident, and it can be entirely manufactured. The

3
1
1
210
Article

The Legible Mind

Minds confabulate. Every one of them, constantly, and none of them can tell. Ask a person why they did something and the answer arrives fluent and confident, and it can be entirely manufactured. The

2
2
5
1,573
Article

The Memory Wars Are Coming

Memory is sticky. Leaving the memory provider means lobotomizing the agent. Inference is commoditizing. Today's frontier model is tomorrow's free tier, and compute providers are locked in a structural

11
8
83
2,238,935
more replies
2/ labs are building memory into the model. chatgpt memory doesn't leave chatgpt. claude can't see across projects. vertex needs gemini to extract. frameworks are building it into the framework.
1
1
603
3/ vector dbs are building it into the database. pinecone just shipped an entirely new product because raw vector search isn't the right shape for memory. they're conceding the layer in public. memory is infrastructure. that's the side that wins.
1
208
Article

Two Times Are Better Than One: Why AI Memory Needs Bi-Temporal Reasoning

Ask an AI agent what it knew last month and most systems can't tell you. They can tell you what they know now. They cannot reconstruct what they believed before yesterday's update overwrote it. The
2
2
139
1/Most AI memory systems flatten everything into one layer. When the agent tells you what it believes, you can't ask why. The chain of reasoning has been compressed away. Here's what memory looks like when you don't do that
CortexDB.ai by creator of Cassandra@cortexdbai
May 26
Article

The Five Memory Layers: Why Flat AI Memory Loses the Plot

When an AI agent tells you what it believes, you should be able to ask why and get a real answer. Today, with most memory systems, you can't. The agent will produce a confident-sounding sentence, and

1
1
1
147
more replies
7/What the structure unlocks: → Evidence traceability: walk a Belief down through Facts → Episodes → raw Events → Safe schema evolution: rebuild derived layers under a better extraction model, from the Events you already have → Genuine temporal reasoning: "What did we believe on day X, as of day Y?"
1
1
1
72
8/Memory is the substrate on which an agent's continuity of self is built. If it can't distinguish observation from inference, can't preserve evidence, and can't reason about time — the agent on top will be confidently wrong in ways its operators can't diagnose. Five layers isn't a feature. It's the minimum.
1
61
Load more