Head of AI Research Product
Via · New York, New York
About this role
**Via — Head of AI Research Product** Via is a global leader in smart transportation, using technology to revolutionize mobility worldwide. This is an opportunity to join a core team reshaping how public mobility operates at scale. You’ll lead the intersection of applied machine learning research and product innovation across Via’s rider and transit operations platform—building the engine that makes the system smarter, more equitable, and more efficient. --- ### What you’ll do - Build and lead Via’s AI research product function, defining the roadmap for how machine learning advances rider experience and transit network performance. - Design and deploy reinforcement learning systems to optimize routing, pooling efficiency, and real-time dispatching—balancing rider satisfaction with vehicle utilization and agency cost targets. - Pioneer AlphaGo-style machine learning for transit: combine deep reinforcement learning with tree search over high-fidelity simulations of real transit networks, training policy/value networks to discover dispatching, pooling, and network-design strategies beyond hand-built heuristics or conventional optimization. - Develop weighted ensemble models and multi-objective optimization frameworks to surface the right tradeoffs between wait times, cost per ride, emissions, and equity of service. - Invent personalization systems that adapt the rider mobile experience in real time (relevant trip options, proactive alerts, and dynamic pricing signals) in close partnership with the rider mobile product team. - Research and productize behavioral incentive models to increase ridership and shift demand to improve aggregation (surge smoothing, loyalty structures, and gamified engagement grounded in behavioral economics and causal inference). - Partner with lighthouse transit agencies to co-develop and validate novel AI approaches, turning field pilots into scalable platform features. - Establish rigorous experimentation infrastructure (A/B testing, contextual bandits, offline policy evaluation) to measure real-world impact on rider behavior and agency outcomes. --- ### What we are looking for - MS or PhD in a quantitative field (machine learning, operations research, statistics, applied mathematics, or related). - Deep hands-on proficiency in ML: reinforcement learning, supervised/unsupervised methods, probabilistic modeling, and optimization. Experience with AlphaGo-style approaches (self-play, Monte Carlo tree search, policy/value networks) and large-scale real-time inference is a strong plus. - Genuine user empathy—thinking carefully about rider and dispatcher experiences alongside model architecture. - Strong outcome focus and clarity about balancing delightful user experience with service KPIs (efficiency and rider experience). - Business sense and product instincts: translating research into shipped products and business outcomes. - Experience partnering with engineering teams to move models from research into production at scale. - Track record working with external customers/partners to iterate on novel solutions in ambiguous, real-world conditions. - Comfort operating across research, product, and commercial contexts (communicating both technical architectures and product strategy). - Passion for building technology that improves public transit for everyone, especially riders in underserved communities. --- ### Compensation & Benefits - Final salary determined by experience, knowledge, and skills. Salary range
Listing freshness
CronJobs last confirmed this listing 1h ago. If its source stops confirming the opening for seven days, this page is removed from active inventory.