Performance Modeling Engineer

Etched was founded on the belief that new hardware will be needed to run human- and superhuman-level artificial intelligence. Our ASIC systems deliver orders of magnitude higher performance than conventional AI chips. Etched is actively developing Sohu, an ASIC exclusively for transformers, which will enable products that would be impossible with GPUs, like real-time video generation models and extremely deep chain-of-thought reasoning. And we're bringing this technology to market in months, not years. Sohu is hard-coded for individual model architectures and only supports transformers, but has an order of magnitude more throughput and lower latency than a B200. 

WHAT YOU’LL DO

• Develop and own SoC functional models from architecture to deployment, including integration with infrastructure, thorough testing, and debugging.
• Collaborate closely with architecture, RTL design, design verification, emulation, and software teams to ensure alignment and performance.
• Innovate tooling and infrastructure to enhance usability and effectiveness for internal and external partners.
• Optimize model and infrastructure performance, ensuring scalability for real-world AI workloads.
• Create software that is well-documented, maintainable, and reusable, ensuring longevity and ease of future development.
• Simulate performance of existing and upcoming state-of-the-art (SOTA) Transformer models on our ASIC architecture vs other alternatives (GPUs, TPUs etc)

WE’RE LOOKING FOR ENGINEERS WITH:

• Expert knowledge in functional modeling of ASICs (TPUs, GPUs, CPUs, or equiv. accelerators)
• Strong command of C++ for modeling with Python familiarity for scripting and automation.
• Insights bridging simulation to analysis of complex systems involving multiple domains (e.g., physical link IO / memory / chip firmware / system software / overall performance).
• High interest in learning transformer architecture, developing insight into AI hardware/software co-design.
• Passion for AI and a desire to push performance and efficiency boundaries.

Machine learning experience is not required. This role will provide opportunities to ramp on relevant ML concepts on the job.

Minimum qualifications:

• Bachelor's degree or equivalent practical experience.