- Leading Data Science in Consumer personalization- Working with 10+ MLEs to optimize our rankers, recommendation system, and incentivizing exploration.

● Earner Guidance- Objective: Enhance the earning and learning experiences of earners through nudges.- Key project: Developed guidance alerts utilizing the Contextual Multi-armed Bandit algorithm(Thompson sampling MAB + XGBoost) to optimize Go Online and Supply Hours.- Impact: Significantly improved marketplace and Earner KPI, validated through A/B experimentation.- Leadership: Spearheaded the integration of data science methodologies into guidance alerts, leading a cross-functional team of engineers and junior data scientists to successful product development.● Research- Objective: Delve into questions of higher scope to build foundational knowledge.- Key project I: Built Glop-based matching solver to simulate the impact of earner-level intervention onmarketplace outcomes, including dispatch completion rate, rider spinner time, and ETA.- Impact: Within months, two P0 work streams have already employed the simulator to expedite headroom analysis.- Key project II: Led pricing strategy for the subscription model to replace per-trip service fee model,factoring individual rationality and incentive compatibility constraints.- Impact: Validated viability through experimentation and rolled out in select markets.● App Platform- Objective: Ensure the health and expansion of the earner app.- Key project: Collaborated cross-functionally with engineering, safety, and marketing teams to integrateour earner app with Apple CarPlay screen.- Impact: Within six months of launch, over 25% of eligible earners in the US adopted this feature. Theproduct’s success led to its receiving the prestigious Earner Magical Experience Award at Uber.

Driver (Earner) Progression Team

Was lead developer for logical optimization engine of SQL processing in Tibero, an RDMBS.Designed architecture of frontend optimization engine for Tibero version 8.