Hardware Performance Modeling Architect
Cerebras is developing a radically new chip and system to dramatically accelerate deep learning applications. Our system runs training and inference workloads orders of magnitude faster than contemporary machines, fundamentally changing the way ML researchers work and pursue AI innovation.
We are innovating at every level of the stack – from chip, to microcode, to power delivery and cooling, to new algorithms and network architectures at the cutting edge of ML research. Our fully-integrated system delivers unprecedented performance because it is built from the ground up for the deep learning workload.
Cerebras is building a team of exceptional people to work together on big problems. Join us!
- You will work with the Hardware and Software teams to analyze and optimize workload performance.
- Develop tools to analyze hardware performance and identify bottlenecks and optimization opportunities.
- Develop performance infrastructure to validate hardware architecture tradeoffs
- Benchmark hardware performance and compare new architectural proposals against existing benchmarks
- Work with SW teams to model expected software performance on new HW architectures
- Develop automation to regress hardware performance
Skills & Qualifications:
- Performance architect, with 10+ years of experience
- Strong programming: C++, Python, multi-thread, multi-process
- Experience with end-to-end workload analysis from low level assembly instruction code to high level distributed algorithms.
- Experience with performance analysison: CPUs, GPUs, TPU, parallel architectures / distributed systems, dataflow / spatial architectures, many-core multi-thread environments
- Strong architecture background and experience building performance modeling infrastructure
- PhD or Master’s degree in Computer Science, Electrical Engineering, or equivalent,
- Focus in computer architecture is desirable
Los Altos, CA or San Diego, CA or Toronto, Canada
- Headquarters/Sunnyvale Office