Lessons from Applied AI: Tesla, Waymo, and Beyond
Aishwarya Jadhav has worked at the forefront of applied AI: from large-scale recommender systems in finance to malaria mapping in Africa, and from self-driving systems at Tesla and Waymo. With a background in multimodal deep learning and computer vision from Carnegie Mellon University, she now leads efforts in gesture understanding and pedestrian semantics at Waymo.
In this conversation, Aishwarya shares what it really takes to bring AI from research to production in safety-critical environments. We’ll discuss the trade-offs between accuracy and performance, how to evaluate perception models for deployment, and what lessons from AI for social good can teach big-tech engineering teams.
We plan to cover:
What it takes to move perception and multimodal models into production
How to evaluate and validate safety-critical AI systems
Where multimodal AI makes a real difference, and where it adds complexity
Lessons from malaria vector mapping for low-resource, high-impact applications
Skills and habits engineers need to thrive in applied AI roles
About the speaker
Aishwarya Jadhav is a Machine Learning Engineer at Waymo, where she leads projects on gesture understanding and pedestrian semantics. She previously worked at Tesla Autopilot AI on multimodal LLMs and video understanding, and at Morgan Stanley on large-scale recommendation systems. Aishwarya holds a Master’s degree in Computer Science from Carnegie Mellon University, where she focused on multimodal deep learning and text information extraction. She has also contributed to social impact projects, including malaria vector mapping in Africa, which helped reduce mosquito populations by 60% at minimal cost.
DataTalks.Club is the place to talk about data. Join our slack community!