Raman Ebrahimi
I am a Ph.D. student in Machine Learning and Data Science at the University of California San Diego, advised by Dr. Massimo Franceschetti. My research interests are in network economics, strategic classification, network games, and generally how we function as a society.Most of what I work on comes down to one question: what do people do when they can only see part of the system they are in? A leaderboard, a feed, or a dashboard decides who you think you are up against, and that changes how hard you push, whether you coordinate with others, and whether you end up outsmarting yourself. I build small game-theoretic models of this, with agents who reason a few steps ahead and see the world through a biased sample of their network, and then check them against user studies. The same thread runs through the rest: how people game a classifier when some features can be faked and some get audited, and how people shade the opinions they report on different platforms, so that the polarization we observe looks worse than it really is. The details are on my Google Scholar.
Currently, as a Data Solutions Engineer at HCM Tradeseal, I build full-stack AI systems, work with customer data, and maintain the WageFinder prevailing-wage data platform (FastAPI, 5.9M records across 55 federal jurisdictions). Until June 2026 I was a Machine Learning Engineer and Quantitative Researcher at MarketCrunch AI, redesigning prediction ensembles for 2,100+ equities, running SHAP-based model diagnostics, and building the MarketPulse signal service and an automated trading system.
On the side I build tools I want to exist: a self-hosted agentic research pipeline on a GPU workstation, an open-source Claude Code research plugin, Antipode (a debate platform), and First Assist, which finds undocumented boulders from LiDAR. See Projects.
Before coming to UCSD, I earned dual Bachelor’s degrees in Industrial Engineering and Physics from Sharif University of Technology in Tehran, Iran. Beyond academics, I’m passionate about how financial markets and complex systems work, I enjoy photography, and rock climbing!
Thanks for visiting my webpage!
Previously:
- Graduate Student Researcher, Multi-agent Intelligence and Decision Systems (MINDS) lab, UCSD (2022–2025): I developed game-theoretic frameworks and ML models incorporating cognitive biases, analyzed Nash equilibrium conditions in multilayer network games to derive policy insights from real-world data, and designed interactive strategy games with ML-driven simulations to study strategic behavior.
- Data Analyst Intern at the American Institute for Behavioral Research and Technology (AIBRT), Dec 2024–Feb 2025: I revived a legacy Python/SQL graphing application for the ‘Generativity Theory’ project, integrating PyTorch and scikit-learn models for user-agnostic behavior prediction (over 90% accuracy in select modes) and building recording and visualization tools for behavioral analysis.
- Data Science Product Manager at the Scientific Association of Industrial Engineering (2019–2021): I led a team of computer science students building a real-time strategy game in Unity, designed the in-game economy, ran market predictions and simulations to test servers and find exploits, and monitored live competition data to detect anomalies.