
Open Vision
Give an agent evidence from video, not a vague summary.
Open Vision turns video into timestamped visual and audio evidence, then returns structured output that an agent can use.
READ PROJECT NOTES →SELECTED WORK / INDEPENDENT PROJECTS
I use small projects to work through questions that are easier to answer in code than in a deck. One gives agents evidence from video. One studies refusal behavior in language models. One proves an identity claim without handing over the document. The last is a research result: changing what a small model can do by editing its weights.

Give an agent evidence from video, not a vague summary.
Open Vision turns video into timestamped visual and audio evidence, then returns structured output that an agent can use.
READ PROJECT NOTES →
Study refusal as a behavior you can measure and intervene on.
Blasphemous is a research toolkit for examining refusal-related representations in transformer activations and testing interventions against an evaluation suite.
READ PROJECT NOTES →
Prove the claim without handing over the document.
ZKYC pairs a Circom zero-knowledge circuit with a fine-tuned vision model, so a browser can prove an identity claim while the document itself never leaves the device.
READ PROJECT NOTES →New abilities by editing existing weights, not by training.
About 3.28M of 873.4M weights changed (0.38%). No training run, no added parameters, no donor model at inference. The math gain repairs thinking-mode looping and default mode costs 2 to 4 points; the write-up lists the limits.
PUSH-T PHYSICS
50% → 89%
DEPTH JUDGMENTS
75% → 83%
MATH-500, THINKING
6% → 53%
NEXT
I am interested in work where software and applied AI have to survive outside a demo.
Surface. Thanks for diving.