About Me
I’m Connor Forsythe, a Decision Scientist at Disney with a Ph.D. in Mechanical Engineering from Carnegie Mellon University. I completed a year-long postdoctoral appointment in CMU’s Engineering and Public Policy department, where my work examined how technology and policy shape transportation choices. I build models that translate complex data into decisions that improve products, operations, and customer outcomes, and I’m passionate about privacy-preserving AI technology. My work combines discrete choice modeling, applied econometrics, and machine-learning workflows to forecast behavior, quantify tradeoffs, and evaluate interventions. I bring a strong programming toolkit—Python, MATLAB, Stata, R, SQL, and Julia—and a track record of delivering analytics that drive measurable impact. I’m always happy to connect. You can reach me at connorrforsythe@gmail.com, connect with me on LinkedIn, or view my publications on this site and my Google Scholar page.
Research
Journal Publications
- Forsythe, C. R., Gillingham, K. T., Michalek, J. J., & Whitefoot, K. S. (2023). Technology advancement is driving electric vehicle adoption. Proceedings of the National Academy of Sciences of the United States of America, 120(23), 1–7. https://doi.org/10.1073/pnas.2219396120
- Forsythe, C. R., Harper, C. D., & Michalek, J. J. (2024). Bringing home the bacon: Estimating willingness to pay for autonomous grocery delivery across U.S. households. Transportation Research Interdisciplinary Perspectives, 26. https://doi.org/10.1016/j.trip.2024.101118
- Forsythe, C. R., Jha, A., Michalek, J., & Whitefoot, K. (2025). Externalities of Policy-Induced Scrappage: The Case of Automotive Safety Inspections. Journal of the Association of Environmental and Resource Economists. https://doi.org/10.1086/739286
- Burns, A., Forsythe, C. R., Michalek, J. J., & Whitefoot, K. S. (2025). Estimating the potential for dynamic parking reservation systems to increase delivery vehicle accommodation. Transportation Research Part A: Policy and Practice, 193. https://doi.org/10.1016/j.tra.2025.104380
- Forsythe, C. R., Gillingham, K. T., Michalek, J. J., & Whitefoot, K. S. (2026). Will pickup-truck buyers go electric? Transportation Research Part D: Transport and Environment, 153, 105170. https://doi.org/10.1016/j.trd.2025.105170
Submitted to Journal
- Koling, A., Armanios, D., Michalek, J. J., Forsythe, C. R., & Jha, A. (2026). Ride-Sharing the Wealth – Effects of Uber and Lyft on Jobs, Wages and Economic Growth. R&R at Nature Cities.
- Vicente, J.P., Forsythe, C. R., Gillingham, K.T., Michalek, J. J., & Whitefoot, K. S. (2025). Automotive Choice Models: A Systematic Review of the State of the Literature. Under review at Renewable and Sustainable Energy Reviews.
Works in Progress
- Forsythe, C. R., Arteaga, C., & Helveston, J. P. (2025). The Mixed Aggregate Preference Logit Model: A Machine Learning Approach to Modeling Unobserved Heterogeneity in Discrete Choice Analysis. [Working Paper]
- Bruchon, M. B., Forsythe, C. R., & Michalek, J. J. (2023). Should Ridesourcing Services Pool More Ride?
- Bruchon, M. B., Forsythe, C. R., Andreasan, C., Whitefoot, K. S., & Michalek, J. J. (2023). Does Congestion Pricing for Uber and Lyft Work? Effects of Chicago’s Downtown Zone Surcharge.
Code & Projects
BonsAI
Project siteBonsAI is a local-first AI desktop app that runs entirely on your own computer. It lets you interact with a modern language model without sending your data to the cloud, without accounts, and without third-party servers in the loop.
The goal is simple: give people the usefulness of AI while keeping control, privacy, and ownership of their data in their own hands. Everything runs locally, and nothing leaves your machine unless you explicitly choose to share it.
Right now, BonsAI is in alpha and supports Apple Silicon Macs only. It’s still early, evolving quickly, and very much a personal, in-progress project.
I built BonsAI because I wanted a tool that felt trustworthy, transparent, and respectful of users — something that worked for me personally before it worked for anyone else.
MyTempo
MyTempo is a precise, programmable metronome for serious musicians. Instead of a static click, it gives you a conductor: build Arrangements that map a full song structure, automate transitions from 4/4 to 7/8, and control exact bar counts for every section.
It supports advanced polyrhythms, custom swing feels, and organized setlists, all in a high-contrast interface built for music stands and dark venues. I built MyTempo because I was tired of subscription-based metronomes being the default; this one is focused, no ads, no clutter, just timing you can trust.