Sotabase

Career

· Member Of Technical Staff, Microsoft AI2025–
· Research Intern Algorithm Development Language Model Development, Google DeepMind2023–2023
· Doctor of Philosophy - PhD, Computer Science, University of California, Berkeley2019–2025
· Research Assistant, Berkeley RISE Lab2017–2019
· Student Researcher Machine Learning (ML) Artificial Intelligence (AI) Distributed Systems, Berkeley RISE Lab2017–2025
· Software Engineering Intern, KATERRA2017–2017
· Bachelor of Science - BS, Electrical Engineering and Computer Sciences, University of California, Berkeley2015–2019

Publications (9)

MoE-Lightning: High-Throughput MoE Inference on Memory-constrained GPUs
International Conference on Architectural Support for Programming Languages and Operating Systems · 2024
46
cited
Managing Bandwidth: The Key to Cloud-Assisted Autonomous Driving
arXiv.org · 2024
3
cited
Context-Aware Streaming Perception in Dynamic Environments
2022
D3: a dynamic deadline-driven approach for building autonomous vehicles
2022
Leveraging Cloud Computing to Make Autonomous Vehicles Safer
2023
Pylot: A Modular Platform for Exploring Latency-Accuracy Tradeoffs in Autonomous Vehicles
2021
Scalable Multi-Domain Adaptation of Language Models using Modular Experts
2024
Time Constraints and Fault Tolerance in Autonomous Driving Systems
2019
TURBO: Utility-Aware Bandwidth Allocation for Cloud-Augmented Autonomous Control
2025
Sotabase
Peter Schafhalter | Researcher Profile | Sotabase | Sotabase