AI Fieldwork is where Zorost evaluates models, agents, and tools on real work, and builds working experiments to probe what the technology can and cannot do yet. Each study is dated, evidence-led, and honest about its limits.
The rest of the AI Lab shows what we build and run in production. Fieldwork is the smaller, sharper habit beside it: structured tests and working experiments that we publish so others can check the method and learn from it. Cutting edge, stated plainly, with the receipts attached.
We test language and domain models on real tasks, record the exact version and window, and report reliability, grounding, and failure modes, not just wins.
We probe tool use, permission boundaries, recovery, and injection resistance. Autonomy is interesting only when it is safe under supervision.
We put AI tools through the work we actually do, and write down where they help, where they slow us down, and where they quietly break.
We build working concepts that push a technique to its edge, then publish the source and the honest engineering account behind it.
A hangar encyclopedia of 265 aircraft types: ICAO designators, published overall size, detailed generated models for headline types, solid schematics for the rest, and class essays for airliners, general aviation, helicopters, and fighters.
A hangar encyclopedia of 265 aircraft types: ICAO designators, published overall size, detailed generated models for headline types, solid schematics for the rest, and class essays for airliners, general aviation, helicopters, and fighters.
A scroll through deep time that makes expansion, star formation, and Doppler shift felt in the body, and argues that AI belongs next to the unsolved problems of the universe, not only next to the solved ones.
yt2textbook prints any YouTube video or playlist as a complete, illustrated textbook: chapters, the screenshots that matter, comprehension checks, a glossary, and a Word export. It runs on your machine with your own key.
We assigned a team of AI agents with separated roles, architects and builders, an adversarial auditor, and a verifier, to turn classical physics and mathematics into a browser instrument: spring lattices, divergence-free flow fields, and more than a hundred figures, from classical curves to sacred geometry and historical ornament, generated live from their equations. This report records what each role carried and what this division of labor opens up for physics, mathematics, and simulation work.
Every study begins with what is being tested, who it is for, and what would count as success, partial success, and failure. We do not start with a preferred winner.
We record versions, configuration, dates, sample sizes, and rubrics. Where results can be reproduced, the method and materials travel with the report.
We use recognized references such as NIST AI RMF, OWASP GenAI guidance, UK AISI Inspect, and HELM to structure testing. Alignment with a framework is not a certificate.
We name what failed with the same clarity as what worked, and we never call a system safe, compliant, or best on the strength of a single test.
We don't pitch slide decks. We show you what we've already built in your domain, then engineer what your mission requires.
