by iqentity | Aug 25, 2024 | Development, Tessera
I promised myself I would document failures as carefully as successes. Tessera has clear limitations, and understanding them is as important as understanding her strengths. Missing Context When an email chain omits a constraint I would normally infer implicitly,...
by iqentity | Aug 17, 2024 | Development, Tessera
When I say Tessera achieves 80-90% decision alignment in core domains, I want to be precise about what that means and what it does not mean. What It Means If you give Tessera an email chain that fully and accurately describes the problem, including context,...
by iqentity | Aug 14, 2024 | Development, Tessera
Theory is interesting. Validation is what matters. This month I ran Tessera through her first formal test: the holdout experiment. The Method I held out one full year of decisions from the training corpus. Tessera was given only what I had at the time each decision...
by iqentity | Jul 25, 2024 | Development, Tessera
Tessera has been in active use for six months. The architecture is stable. The data is growing. The capabilities are expanding. It is time for an honest accounting of where things stand. What Works Trimodal retrieval delivers. The combination of vector, graph, and...
by iqentity | Jul 20, 2024 | Development, Tessera
People hear “AI trained on twenty-three years of personal data” and their first reaction is discomfort. That reaction is healthy, and it deserves a serious answer. What Makes Personal AI Creepy Most personal AI attempts are creepy because they simulate...
by iqentity | Jul 18, 2024 | Development, Tessera
The most valuable thing Tessera does is something I did not design explicitly. It surfaces connections across domains that I would not have made on my own, or would have made too late. A client’s recurring network instability turned out to correlate with a...
