SCANOVICHApplied AI research lab
Independent applied AI research lab · since 2024

AI that a person
can sign.

Fig. 01 · The lab, mapped

Five lines of research, eleven projects, and the evidence behind them.

Every figure on this site links to where it can be checked, with the date it was measured. What we have not measured is not here.

Research linesProjectsEvidence Documents that carry liability 01 Evidence and authority 02 Retrieval and memory 03 The agent-readable web 04 Private, local AI 05 Trade-document engine prototype Document cross-check prototype Services-BGE released Precedent released VoiceToText released Agent Web Lab research EvidenceGene Court experiment Call analytics prototype Private AI workspace prototype Document-intelligence core prototype Evidence-bound decisions research 92% 2026-11-11 6 of 6 2026-11-11 0.861 2026-11-11 0 2026-11-12 293 of 320 2026-11-12 7 of 20 2026-11-12 v2.2.0 · MIT 2026-11-11 Method + results 2026-08-10 Accuracy report 2026-06-13 Demo video 2026-06-13 Reference stack 2026-07-17 Reference stack 2026-07-17
> point at a nodeEach research line leads to the projects that test it, and each project to the evidence behind it.
92%

of questions get the right passage ranked first by search with reranking. Plain text search manages 38%.

MEMORY-BENCH.md · 2026-11-11 ↗
6 of 6

times a planted false note wins without trust tiers. With reviewed files preferred, it never does.

MEMORY-BENCH.md · 2026-11-11 ↗
0.861

recall at rank one with the better of two rerankers on the Russian demo set. The other reaches 0.750.

COMPARISON.md · 2026-11-11 ↗
§ 1The question

A language model will answer anything.

In work that carries liability, an answer is worth something only when a person with authority can sign under it. That takes a source for every value, a rule for every decision, and a system that stops when it does not know. The lab studies where that line is, one domain at a time.

§ 4For agents

Written for people and for their agents.

One source of facts. People get pages. Agents get the same facts as plain text and structured data.

$ curl scanovich.ai/llms.txt 200 a short map of the site $ curl scanovich.ai/ai/catalog.json 200 11 projects · 5 lines · 12 evidence records, each with source and date
How an agent should read this site →
§ 6The lab

One researcher. A governed team of agents.

Aleksandr Mordvinov

Aleksandr Mordvinov

Founder · Principal researcher

More than fifteen years in work where documents carry liability: import and certification of electronics, regulatory permits, customs clearance. Since 2024 he has built the lab’s systems himself.

Each agent in the lab owns one zone. They share one memory, where reviewed facts outrank unreviewed notes. Nothing leaves in the lab’s name without his signature.

About the lab →