URBAN MOBILITY SCIENTIST
I work with transport agencies to evaluate major projects under uncertainty, using real-world data, agent-based models, AI and scenario analysis. I also lecture to companies on leveraging AI tools to transform complex challenges into streamlined workflows.
Current collaborations include the Ministry of Transport's Strategic Planning and Innovation Division, Netivei Ayalon, Netivei Israel, the Jerusalem Master Transportation Team, and the Central Bureau of Statistics (CBS).
TRUSTED BY AND COLLABORATING WITH INDUSTRY LEADERS
With a PhD from Tel Aviv University, I bridge the gap between theoretical algorithms and concrete urban reality. My work focuses on Robust Policy Evaluation—ensuring that billion-dollar decisions handle uncertainty. I also lecture to companies on AI-native workflows, showing teams how to leverage cutting-edge AI tools to transform complex challenges into elegant solutions.
Core technologies used for simulation & analysis:
"Combating Congestion: Robust Transportation Policy Evaluation"
"Evaluating the Impacts of Dedicated Bus Lanes on Urban Traffic with an Agent-Based Model"
Strategic planning for the Ministry of Transport. Harnessing massive API datasets to visualize real-time congestion and optimize traffic count distribution.
Role: Method Lead, Technical Advisor, Co-PI
Budget Responsibility: Lead project up to 15 million (Google Project).
Evaluated "Carrot and Stick" strategies for Jerusalem. Findings showed how congestion pricing stabilizes the impact of Shared Automated Cars on public transport usage.
View Simulation →Rigorous "blind" reconstruction of the Beer Sheva traffic model results to validate the integrity and accuracy of the simulation for government approval.
View Methodology →Pioneering AI-Native Development using Claude, Gemini, and Codex to accelerate simulation pipelines and automate complex spatial SQL queries.
Learn More →
Agent-based evaluation of congestion charges and parking prices in central Jerusalem. MATsim scenarios show how fees reduce congestion and emissions and encourage travelers to switch from private cars.
Parallel framework for large-scale urban traffic simulation in MATsim. Automatically clusters traffic to partition the network, balance cores, and reduce synchronization, delivering faster simulations on real road networks.
Studies MATsim downscaling by comparing full and sampled populations in Sioux Falls. Shows which reduced agent shares preserve key traffic statistics and where further scaling distorts network dynamics.
Open to consulting on national and metropolitan models, data infrastructure for mobility analytics, and research collaborations.
