MLBytes: Synthetic Data Simulation for Autonomous Driving

Engineers and researchers working on autonomous driving are developing complex models to put self-driving cars on the road, but these models are hard to test. In particular, edge cases, by definition, are difficult to find, since they are rare. This talk describes how simulation and re-simulation using synthetic data are being used to test models and improve the generalization of various components of the autonomous driving stack, including planning and perception. Applied Intuition provides software infrastructure to safely develop, test, and deploy autonomous vehicles at scale, including using simulation and synthetic data.

Speaker Bios:
Shrey is a software engineer at Applied Intuition, working on search and APIs for autonomous driving simulation. He graduated from Duke (’20), where he studied computer science and statistical science. Before joining Applied, Shrey did research & software engineering at Google on the Search team, quantitative research at a hedge fund on the equities group, and data science & software engineering at several startups. He also co-founded Duke Undergraduate Machine Learning during his time at Duke, and led the planning of the first Duke Datathon!

Katherine is a software engineer at Applied Intuition, working on cloud usability and infrastructure for autonomous driving simulation. She graduated from Duke (’19), where she studied electrical & computer engineering and computer science. Before joining Applied, Katherine was at Sidewalk Labs (part of Alphabet/Google) as an engineer working on urban planning tools. She also did software engineering at Google, and at Leanplum as part of the Kleiner Perkins Fellows program.