"Company brain" is currently three different products wearing one name.
Enterprise search indexes documents and finds them. A knowledge base stores documents and organises them. Both are genuinely useful, and neither can answer "how does the customer named in this doc relate to
The graph-engineering conversation and the company-brain conversation are the same conversation, and almost nobody is treating them that way.
Agent graph: nodes are agents, edges route work, shared state flows along the edges. The part that gets glossed over every time is where
We ran LongMemEval-s. 90.79% overall. The two numbers we actually cared about: Knowledge Update 97.43%, Temporal Reasoning 90.97%. Here is why those two, and where we are still soft.
A customer moved last week. Your DB now holds their old address, their new address, two orders shipped to the old one, and a support thread about the move.
Every row is relevant. One address is correct.
Similarity search gets you to the right topic. It cannot tell you which
