about the lab

Good questions.
Experiments you can inspect.

plicara is an independent AI research lab. We study how systems behave, what makes them useful, and how to tell when they work.

Start with a concrete question and a simple baseline. Build an experiment, inspect the failures, and revise the explanation when the evidence changes. State the limits alongside the result.

Research articles, benchmarks and small tools that help others investigate the same questions. We aim to publish the evidence needed to check a claim, and identify gaps when that evidence is incomplete.

I am Adrian Tame, and I run plicara. You can find me in these places.