Deepak P- Email & Phone Number

United States

Mountain View, California, United States

- Developed requirement-driven full-system tests to verify fault-tolerant LiDAR and Perception software for Waymo’s ride-hailing and trucking platforms.- Extended a fault-injection framework to support scaleable and automated LiDAR software testing.- Built software tools to improve the clarity and traceability of fault-injection test artifacts.- Developed simulation tests using data mutation/augmentation to verify LiDAR data quality issues are detected.- Signed-off LiDAR software for key program milestones, including the launch of Waymo’s fully autonomous ride-hailing service in Phoenix and San Francisco.

Mountain View, California, United States

Sensor calibration and validation, and fleet scaling.

Claremont, California, United States

Worked on a team of 5 to design and build an mmWave radar perception system to detect and track obstacles in real-time, for use on indoor industrial AGVs.- Senior Capstone project at Harvey Mudd College, sponsored by Ningbo Techmation Co. - Team lead for the Spring semester.- Implemented a multi-object detection and tracking algorithm with the mmWave radar, running in real-time on an Nvidia Jetson TX2. The algorithm used DBSCAN, a constant acceleration model, gating and data association, and an Extended Kalman Filter.- Characterized tracking system performance and latency. - Tested and characterized performance of a short range, high resolution mmWave radar (TI IWR1642 and IWR1443) in detecting objects of various sizes and radar cross-sections across different configurations and indoor environments. - Wrote a Python tool to parse data packets streaming from the mmWave radar, in real-time.- Integrated system on an indoor robotic platform with ROS. The robot could navigate autonomously while avoiding static and dynamic obstacles by purely using our mmWave radar perception system, and an occupancy grid and cost map. - Presented work at project sponsor HQ in Shanghai, China.

Mountain View, California, United States

Project Taara - wireless, high-throughput optical communication.- Built and tested the hardware and software for a bench-top atmospheric channel emulator, used to characterize the optical communication system performance. Presented findings to senior leadership.- Setup and tested an ASIC for a novel modulation scheme to be used in future systems.- Programmed the data-processing pipeline for a system telemetry and performance visualizer, to improve and expedite system monitoring.