alt text

Hi there! I’m Dan.

I’m a researcher, computer scientist, and mathematician based out of New York City. I currently work at Jane Street Capital, a quantitative trading firm.

I really like building things–the entire process, from ideation to implementation. I think we have an obligation to leave the world better than we found it, and I believe my best way of doing so is through technological innovation and research. As of late, I’ve been most interested in non-convex optimization, computational physics, AI alignment, and natural language processing.

I consider myself a life-long learner–I read textbooks and research papers in my free time, and I’m making an effort to consolidate my notes here. I’m passionate about effective technical communication and writing; you can find my blog here.

If you’d like to contact me, I can be reached at: dan DOT dipietro DOT pub @ gmail DOT com


For detailed information on my research, please see the appropriate page. Also, my Google Scholar profile is available here.

  • DiPietro, D. M., & Zhu, B. (2022). Symplectically Integrated Symbolic Regression of Hamiltonian Dynamical Systems. arXiv:2209.01521. Paper.
  • DiPietro, D. M., & Hazari, V. D. (2022). DiPietro-Hazari Kappa: A Novel Metric for Assessing Labeling Quality via Annotation. arXiv:2209.08243. Paper.
  • DiPietro, D. M. (2022). Quantitative Stopword Generation for Sentiment Analysis via Recursive and Iterative Deletion. arXiv:2209.01519. Paper.
  • DiPietro, D. M., Hazari, V. D., & Vosoughi, S. (2022). Robin: A Novel Online Suicidal Text Corpus of Substantial Breadth and Scale. arXiv:2209.05707. Paper.
  • DiPietro, D. M., Xiong, S., & Zhu, B. (2020). Sparse Symplectically Integrated Neural Networks. Advances in Neural Information Processing Systems. Paper.
  • Deng, Y., Zhang, Y., He, X., Yang, S., Tong, Y., Zhang, M., DiPietro, D. M., & Zhu, B. (2020). Soft Multicopter Control using Neural Dynamics Identification. Conference on Robot Learning. Paper.
  • Fleiss, A., Cui, H., Stoikov, S., & DiPietro, D. M. (2020). Constructing Equity Portfolios from SEC 13F Data Using Feature Extraction and Machine Learning. Journal of Financial Data Science, 2(1), 45-60. Paper.
  • DiPietro, D. M. (2019). Alpha Cloning: Using Quantitative Techniques and SEC 13F Data for Equity Portfolio Optimization and Generation. Journal of Financial Data Science, 1(4), 159-171. Paper.

Software Projects

For detailed information on my software projects, please see the appropriate page. Most of my research has involved exciting and novel computational implementations; these are recorded on the research page.

  • Fairer Features: ML pipeline that uses computer vision and large language models to extract latent demographic and portrayal information from image datasets. A novel optimization approach re-weights the images to reduce any present biases. Interviewed for YCombinator 2023W.
  • DSRPytorch: From-scratch PyTorch implementation of Deep Symbolic Regression. Implements RL policy gradients, batching of variable-length sequences, and a sequence-to-PyTorch transpiler (allowing models to write PyTorch code).
  • GPU Optimization Code: Implemented a conjugate gradient solver using CUDA. Via some clever computational tricks, my implementation beats a naive GPU implementation by 10x.
  • D2 BioSoftware: Computer vision project that detects and counts the number of bacterial colonies present on a petri dish. Implemented using OpenCV and Electron; used by the McGowan Institue of Regenerative Medicine at the University of Pittsburgh.


  • I’m an avid botanist and carnivorous plant collector, specializing in the genera Nepenthes, Heliamphora, and Drosera. I write about my plants here and have published a book on them, available here.
  • I really like the outdoors, especially skiing and mountaineering. I’m slowly working my way through climbing all peaks in Colorado above 14,000 feet (affectionately called the 14ers). I did 11 last summer!
  • I enjoy playing jazz piano and drums.