Data Parallel Accelerator Performance Intern
About the job
SummaryBy Outscal
Our mission is to create computing platforms (HW/SW co-design) that will transform the industry with the most advanced technologies. As a DPA performance intern, you will be given a project to work on SOC-level performance per watt improvement through memory management innovations. You will be working with the internal SW (eg. OS, Kernel, Framework) and Silicon (eg. RTL, Power, Perf) team members
Requirements
- Knowledge in one or more of the following areas, computer architecture , performance modeling, and analytical model
- Knowledge and experience with common LLM (Large Language Model) workloads.
- Proficiency in C or C++, and scripting languages such as Python.
- Experience with high-level simulators for performance or power estimation is a plus.
- Knowledge in server-class GPU/ML architecture is a plus.
Responsibilities
- Responsible for an analytical model implementation of LLM inference and training memory usage
- Responsible for running the performance simulation to extract the workload's characteristics such as memory footprint and bandwidth requirement.
- Responsible for evaluation ideas for performance improvement
Minimum Education & Experience
Current EE or CS master or Ph.D students with computer architecture backgrounds
About The Company
Similar Skill Jobs
Quebec, Canada (On-Site)
Shanghai, China (On-Site)
British Columbia, Canada (On-Site)
England, United Kingdom (On-Site)
Telangana, India (On-Site)
Remote, Us Remote, Canada (Remote)
England, United Kingdom (Hybrid)
Austin, Tx Bay Area, Ca Carlsbad, Ca Los Angeles, Ca (On-Site)
İstanbul, Türkiye (On-Site)
Karnataka, India (On-Site)
Jobs in Hsinchu, Hsinchu City, Taiwan
Hsinchu City, Taiwan (On-Site)
Hsinchu City, Taiwan (On-Site)
Hsinchu City, Taiwan (On-Site)
Hsinchu City, Taiwan (On-Site)
Hsinchu City, Taiwan (On-Site)
Hsinchu City, Taiwan (On-Site)
Hsinchu City, Taiwan (On-Site)
Hsinchu City, Taiwan (On-Site)
Hsinchu City, Taiwan (Hybrid)
Hsinchu City, Taiwan (On-Site)
