Computational neuroscience · Bayesian inference · Quantitative research
I am a Computational Neuroscience Ph.D. researcher at Albert Einstein College of Medicine, with a B.S. in mathematics and a minor in applied mathematics and statistics. My primary focus is computational neuroscience and probabilistic modeling, with applications to quantitative research and market microstructure.
Academic research · Market microstructure · Open source · Email
Published work: PLOS Computational Biology (2025) · Journal of Vision (2022)
I use Bayesian models to understand how sensory uncertainty, prior experience, and response mapping shape perception and decisions.
- Co-first author, PLOS Computational Biology (2025): explained response-range-dependent heading biases with an efficient Bayesian observer and a perception-to-action mapping.
- First author, Journal of Vision (2022): separated perceptual and postperceptual contributions to serial dependence in heading judgments.
Bayesian Heading Observer — Python demo and original MATLAB code
85.8 million order messages · Python + C++20
Reconstructed limit order books for five Warsaw Stock Exchange equities and evaluated forecasts, crossing costs, and conditional passive fills. Positive prediction scores did not survive visible spread costs—a central finding of the historical study. The C++ replay backend matched Python outputs exactly and ran 8.98× faster on the fixed replay benchmark.
| Project | Contribution | Public record |
|---|---|---|
| PMXT | Prediction-market SDK compatibility and optional order parameters | Merged: #1064, #1065, #1290 |
| Filecoin Lotus | CLI diagnostics when a flag overrides the configured API address | Merged: #13670 |
| ColaMD (collaboration project) | Search and LaTeX support | Merged: #14 |
Built an open-source CodexBar + Subrouter integration for monitoring multiple Codex subscription accounts on macOS. The optional native integration adds quota bars and manual subscription-account selection for new chats; routing is handled by Subrouter.
Product website · Source & installation
Other tools: taskdone-runner, background-task notifications and review tracking; HTML report skill, readable research reports with structured evidence.
My independent trading work spans prediction-market execution and cross-market relative value, with attention to settlement, hedging, inventory, and financing constraints.
Tools: Python · C++ · SQL · NumPy / SciPy / pandas · PyTorch · MATLAB
Email · Ph.D. expected December 2027