(Left) Adi Tummala and (Right) Yu-An Su
Adi Tummala
Background: Adi is a Computer Vision master’s student at Carnegie Mellon University with a strong foundation in software engineering and AI. He recently worked at various startups in the luxury flight management and legal funding industries. Prior, Adi interned at NASA Ames Research Center, contributing to open-source ProgPy software and exploring generative AI for service-based systems. He is currently interning at Google, working on agentic systems built on their intranet, MoMA.
Expertise: Deep learning and LLM systems, computer vision, cryptography, software development, and autonomous systems.
Project Responsibilities: Joint responsibilities shared with Yu-An, Adi researches and validates current frontier models for depth estimation. Currently he is working on various fusion-based methods for reconstruction, on sparse, or complete removal, of lidar sensor. Adi is working on curating a dataset for the fine-tune runs over SOTA monocular depth models given misaligned lidar data.
Yu-An Su
Background: Yu-An is a Computer Vision master’s student at Carnegie Mellon University’s Robotics Institute, with a strong foundation in perception and autonomous systems. He has conducted research on sensor fusion for autonomous racing platforms. Before CMU, he was a research assistant at Academia Sinica, focusing on Vision Language Model training and alignment. He is currently interning at Zoox, working on perception for autonomous vehicles.
Expertise: Metric depth estimation, sensor fusion, computer vision, autonomous vehicle perception, and robotics software development.
Project Responsibilities: Joint responsibilities shared with Adi, Yu-An researches and validates current frontier models for depth estimation. Yu-An is developing alignment methods to handle LiDAR and RBG imagery timestamp mismatches caused by hardware jitters. Yu-An coordinated and setup small GPU cluster over 4 NVIDIA TITAN Xps.