CV
Download a more complete PDF here
Professional Experience
- AI Verification Engineer (Data Scientist + Software Engineer) - Zoox (2022-Present)
- Conducted applied research on a methods team, developing and deploying novel black-box safety quantification techniques for the AI driving stack. Built tools, automated processes, and data products using Python, Databricks, and PySpark to support autonomous driving evaluation, contributing to 9 patent applications (2 granted)
- Designed a data-collection strategy, developed fleet simulations, and built a test asset orchestration system that enabled scaling and release of the Zoox service to the public. Drove continuous improvements resulting in over 99% uptime
- Decomposed complex problems into known statistical problems, developed prototypes, and deployed tooling to orchestrate the test vehicle fleet and deliver $25M+ of log data for safety validation and AI development
Skills
- Probability & Statistics
- Bayesian inference, Uncertainty quantification, Experimental design, Sensitivity analysis
- Research
- Design of Experiments, Statistical modeling, Data Science, Data visualization, Scientific computing
- Programming
- Python, Bash, Linux/Unix
- Developing: SQL, Rust
- Author of the open source maxpro Rust crate and Python package
- Other: Code review, Software testing, LaTeX, Markdown
- Tools
- Data Science
- Databricks, PySpark
- scikit-learn, Pandas, NumPy, SciPy, Matplotlib, Seaborn, SQLAlchemy, SALib
- MCMC
- Software
- Git, GitHub, GitHub Actions
- Jupyter, Bazel, Airflow, Slurm
- PyO3, Rayon
- Wolfram Mathematica
- Flask, FastAPI
- Teaching
- Pedagogical development for flipping a premier introductory physics course at McGill
- Lab report and exam marking
- Preparing tutorials
- Mentoring students by leading help sessions and producing reference materials (video)
Education
- Ph.D. in Physics - McGill University (2022)
- Thesis: “Quantification of the Quark-Gluon Plasma with Statistical Learning”
- M.Sc. in Physics - McGill University (2018)
- B.Sc.(Hons.) in Physics, German Studies minor - The College of William & Mary (2016)
Research Experience
- Graduate Research Assistant - McGill University (2016-2022)
- Trained, validated, and deployed models for dimensionality reduction, regression, and statistical inference on several hundred gigabytes of high-dimensional data
- Improved computational study design performance by over 50% for model prediction, sensitivity analysis, and uncertainty quantification using advanced statistical algorithms for design spaces robust to inactive dimensions
- Executed and managed 1,000,000+ simulations and 500+ CPU years of computing resources valued at over $70,000 at national High Performance Computing facilities
- Developed, validated, and automated C++ and Python scientific-computing tools for data analysis, streamlining the workflow and reducing manual job orchestration
- Contributed to developing and validating statistical analysis for the most advanced C++ framework of a sophisticated nuclear physics environment (jetscape.org)
- Undergraduate Student Researcher - The College of William & Mary (2013-2016)
- Undergraduate honors thesis using computational methods (Python, Fortran) to probe Big Bang Nucleosynthesis for limits on Beyond Standard Model physics. This work was performed under the direction of Andre Walker-Loud.
- Stirling Cycle Analyst for Nuclear Space Power Applications - NASA Glenn Research Center (2015)
- LERCIP Intern with the Thermal Energy Conversion Branch working to improve model fidelity and performing simulations for the Test Demonstration Unit (TDU) for the pre-testing Test Readiness Review. Additional work was done to optimize Stirling engines to map power density-efficiency space and customize the piston-displacer waveform. Future applications of this work include deep space and space exploration applications as passive power units
- REU Student - Texas A&M University and Cyclotron Institute (2014)
- National Science Foundation (US)-funded studentship under Ralf Rapp studying hot and dense hadronic matter, resulting in a state-of-the-art parametrization of thermal photon production in hadronic matter.
Publications
M. Heffernan, C. Gale, S. Jeon, and J.-F. Paquet, (2023). "Early-times Yang-Mills dynamics and the characterization of strongly interacting matter with statistical learning." arXiv preprint 2306.09619
M. Heffernan, C. Gale, S. Jeon, and J.-F. Paquet, (2023). "Bayesian quantification of strongly-interacting matter with color glass condensate initial conditions." arXiv preprint 2302.09478
M. Heffernan (2021). "How about that Bayes: Bayesian techniques and the simple pendulum." arXiv preprint 2104.08621
M. Heffernan, S. Jeon, and C. Gale (2020). "Hadronic transport coefficients from the linear sigma model at finite temperature." Phys. Rev. C. 102, 034906.
M. Heffernan, P. Banerjee, and A. Walker-Loud. (2017). " Quantifying the sensitivity of Big Bang Nucleosynthesis to isospin breaking with input from lattice QCD." [arXiv:nucl-th/1706.04991].
M. Heffernan, P. Hohler, and R. Rapp. (2015). "Universal parametrization of thermal photon rates in hadronic matter." Phys. Rev. C. 95(027902).
Talks
November 30, 2023
Talk at APS DNP Fall Meeting 2023, Waikoloa, Hawaii
April 18, 2021
Talk at April Meeting 2021, Virtual
January 11, 2021
Talk at Initial Stages 2021, Rehovot, Israel (Virtual)
April 10, 2020
Talk at Duke University QCD Group Seminar, Durham NC, USA
October 15, 2019
Talk at APS Division of Nuclear Physics Fall Meeting 2019, Washington, D.C.
Teaching
Service and leadership
- Organizing Committee Member, McGill Physics Hackathon (2018-2022)
- VP Communications, McGill Graduate Association of Physics Students (2017-2019)