Software Engineer - GPU reliability

Hudson River Trading•New York City, NY

23h•$200,000 - $300,000

About The Position

Hudson River Trading (HRT) is seeking a Software Engineer focused on GPU reliability to join our Systems Development team. The Systems Development team builds and maintains the platform that is shared by all Systems teams to provision, monitor, and manage HRT’s server and network infrastructure. In this role, your main focus will be to develop tools in Python to analyze the performance of GPU hardware and build creative solutions to improve observability, reliability, and efficiency of the fleet. You’ll work closely with other engineering teams to deeply understand research and trading workflows and ensure that GPU infrastructure is utilized optimally. Strong Python skills and development experience are required, along with Unix experience and a background of managing GPU hardware at scale. Responsibilities This role offers a unique opportunity to make a significant impact on a critical part of our existing and growing infrastructure.

Requirements

BS and/or MS in computer science or a related field
2+ years of relevant experience, including programming in Python and managing GPUs
Experience using automation to solve problems and improve process efficiency
Experience working with, troubleshooting, tuning, and deploying various types of GPU hardware
Strong grasp of computer science fundamentals and software design patterns
Solid understanding of Linux/UNIX operating systems
Familiarity with open-source software
Ability to debug and analyze problems quickly
Skilled at balancing multiple tasks while maintaining meticulous attention to detail
Ability to operate effectively as a team player and also work independently
Ability to learn at a fast pace and apply new skills effectively

Nice To Haves

Understanding of Debian operating system
Familiarity with systems configuration management and monitoring technologies
Familiarity with continuous integration and continuous deployment tools and processes
Understanding of networking protocols

Responsibilities

Building and maintaining tools and software features to automate systems engineering workflows related to GPU management, monitoring, metrics collection, maintenance, and network configuration
Troubleshooting software and hardware bugs on a fleet of GPU devices, including application, network, operating system, and/or kernel issues
Working across HRT’s engineering teams to tune workloads and processes to use GPUs more efficiently
Analyzing GPU job statistics to identify trends and areas for improvement