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.
Technical Program Manager
Etched is seeking a highly skilled and motivated Technical Program Manager to join our dynamic team. This strategic leader will drive collaboration among various internal and external stakeholders, including our platform solutions, ASIC silicon design team, software organization, and external vendors. In this critical role, you will ensure the successful delivery of our AI datacenter hardware projects, identify risks, and optimize engineering and operational efficiency.
Representative Projects
- Collaborate with internal stakeholders, including platform solutions, ASIC design, and software teams, to ensure alignment and effective communication throughout all phases of the project development.
- Oversee external vendors responsible for specific design components by managing timelines and deliverables while ensuring adherence to quality standards.
- Identify and monitor critical paths and potential risks in engineering processes, collaborating with engineering teams to develop mitigation strategies and contingency plans.
- Manage the planning, scheduling, and execution of complex technical projects, ensuring timely delivery within scope and budget.
- Communicate regularly with executive teams and engineering leaders on project status, key risks, and improvement opportunities, fostering a culture of accountability and transparency.
- Continuously assess and refine program management processes to enhance engineering efficiency and operational performance, driving the implementation of best practices.
- Establish key performance indicators (KPIs) to monitor project progress and success, providing data-driven insights for decision-making and strategic planning.
- Leverage knowledge of hardware, ASIC design, and software integration to effectively communicate and resolve issues across engineering disciplines.
You might be a good fit if you have
- Bachelor’s degree in Engineering, Computer Science, or a related field; advanced degree preferred.
- 5+ years of experience in technical program management, preferably within the hardware, semiconductor, or AI technology space.
- Proven track record of managing complex, cross-functional projects with multiple stakeholders.
- Strong understanding of ASIC design, hardware platforms, and software integration.
- Exceptional communication, organizational, and leadership skills.
- Experience working with external vendors and managing supplier relationships.
- Ability to proactively identify and mitigate risks, with a keen eye for process improvement opportunities.
- Familiarity with program management tools and methodologies (e.g., Agile, Waterfall, MS project, Smartsheets).
- Ability to work effectively in a fast-paced, dynamic environment.
We encourage you to apply even if you do not believe you meet every single qualification.
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
What We Do
By burning the transformer architecture into our chips, we’re creating the world’s most powerful servers for transformer inference.
