Compensation Overview

About Etched

Etched is building AI chips that are hard-coded for individual model architectures. Our first product (Sohu) only supports transformers, but has an order of magnitude more throughput and lower latency than a B200. With Etched ASICs, you can build products that would be impossible with GPUs, like real-time video generation models and extremely deep chain-of-thought reasoning.

Linux Driver Engineer, Networking

We are looking for a Linux driver engineer with experience of high-speed data center networking for scalable parallel AI infrastructure. The ideal candidate will be able to define a system scale-out plan for the design and implement Linux drivers for networking devices using RoCEv2.

Representative projects:

• Develop high-speed Linux networking device driver and optimize system performance for large AI/ML use-cases.
• Collaborate closely with system engineers to select the right hardware and software stack.
• Build prototype systems to validate functionality and performance.
• Engage with customers to support deployment of AI inference systems in their own environment.

You may be a good fit if you have:

• 5+ years of experience in Linux driver development.
• 2+ years of experience with Infiniband/RoCEv2 development.
• Ability to learn quickly and have an open-minded mindset.
• Proficiency in C/C++ programming for Linux driver.
• Experience with debugging issues involving Ethernet Switches/Routers and network protocol.
• Knowledge of Infiniband/RoCEv2/RDMA/PCIe/CXL.

Strong candidates may also have experience with:

• Experience with AI infrastructure in large data-centers.
• Experience with multi-system NCCL programming.
• Experience with network accessing patterns with AI inference accelerators.
• Experience with Bazel/Blaze/Buck build system.

How we’re different:

Etched believes in the Bitter Lesson. We think most of the progress in the AI field has come from using more FLOPs to train and run models, and the best way to get more FLOPs is to build model-specific hardware. Larger and larger training runs encourage companies to consolidate around fewer model architectures, which creates a market for single-model ASICs.

We are a fully in-person team in Cupertino, and greatly value engineering skills. We do not have boundaries between engineering and research, and we expect all of our technical staff to contribute to both as needed.

Benefits:

• Full medical, dental, and vision packages, with 100% of premium covered, 90% for dependents
• Housing subsidy of $2,000/month for those living within walking distance of the office
• Daily lunch and dinner in our office
• Relocation support for those moving to Cupertino