• Generative AI• ML-powered product features• Search• Performance• Reliability

Computer vision, AI infrastructure, MLOps

Realtime high-throughput ML systems for Adtech

• Built supervised machine learning models to generate predictions for over 8 million users to support a new commuter product.• Performed financial analysis and drove key product decisions to develop a viable subscription business model for Uber.

• Leveraged data to analyze and optimize Uber's driver and rider growth funnels in collaboration with the relevant Product and Marketing teams• Designed experiments and interpret the results to draw detailed and actionable conclusions.• Generated ideas for exploratory analysis to shape future projects and provide recommendations for actions.• Performed time-series analyses, hypothesis testing, and causal analyses to statistically assess relative impact and extract trends.• Built models to enhance understanding of user behavior and predict future performance of cohorts.• Created dashboards and reports to regularly communicate results to executive stakeholders and monitor key metrics.

Neural networks before they were cool applied to genomic datasets