Thermal Engineer, Rack Level

Job Summary
We are seeking a Thermal Engineer, Rack Level to own the thermal design and validation of Etched’s L11 rack system: the liquid-cooled racks that carry our inference hardware into customer datacenters. You will take the rack from design through deployment. That means defining how it is filled, commissioned, and brought online, validating the reference CDU systems it runs on, and serving as the technical voice on facility integration for our go-to-market team and customers.

This role fits someone who has commissioned liquid cooling systems in a real datacenter and knows what goes wrong between the spec sheet and the white space. You will work closely with our tray- and module-level thermal engineers, with the mechanical, electrical, and systems teams, and with CDU and liquid-loop suppliers.

Key Responsibilities

• Own thermal design and validation of the L11 rack system, including the rack-level liquid loop, manifolds, quick disconnects, and interfaces to facility cooling..
• Develop and document deployment, fill, and commissioning processes for liquid-cooled racks, from first fill and leak checks through bring-up at customer sites.
• Validate reference CDU systems. This includes writing and executing test plans in the lab and in the field, and qualifying CDU partners on thermal performance, reliability, and controls.
• Characterize rack-level thermal and hydraulic performance (flow distribution, pressure drop, supply and return temperatures) and use the results to drive design decisions.
• Define and validate cooling-loop control strategies and how they interact with datacenter facility control systems.
• Support GTM and customers as the technical point of contact on facility integration questions, such as facility water requirements, CDU selection and sizing, secondary loop design, and water quality.
• Partner with tray and module (L6–L10) thermal engineers so that server-level designs meet rack-level and facility constraints.
• Create and maintain rack thermal specifications, commissioning procedures, validation reports, and customer-facing facility requirements documents.
• Investigate and resolve thermal and fluid-related issues in the lab and in deployed systems.

You may be a good fit if you have

• 5–10 years of experience in thermal or mechanical engineering with direct, hands-on CDU and datacenter liquid cooling experience.
• Experience installing, commissioning, or validating liquid cooling systems at datacenter scale.
• Experience with CDU validation and commissioning, including flow, pressure, temperature, and controls testing.
• Strong working knowledge of datacenter cooling loops (facility water and technology cooling system / secondary loops) and the control systems that run them.
• Comfort working directly with customers, facility teams, and suppliers to resolve integration issues.
• A Bachelor’s degree in Mechanical Engineering, Thermal Engineering, or a related field.

Nice to have

• Experience deploying rack-scale direct-to-chip liquid cooling for AI or HPC systems at a hyperscaler, server OEM, or CDU vendor.
• Familiarity with industry guidelines for datacenter liquid cooling, such as OCP and ASHRAE TC 9.9.
• Experience with 1D flow network or system-level thermal/hydraulic modeling tools (e.g., Flownex, MacroFlow).
• Knowledge of coolant chemistry, filtration, leak detection, and fluid integrity testing.

Benefits

• Medical, dental, and vision packages with generous premium coverage
• $500 per month credit for waiving medical benefits
• Housing subsidy of $2k per month for those living within walking distance of the office
• Relocation support for those moving to San Jose (Santana Row)
• Various wellness benefits covering fitness, mental health, and more
• Daily lunch and dinner in our office
• Unlimited compute budget subject to ROI justification

Compensation Range: $200,000 - $350,000 with significant equity.

How we’re different
Etched believes in the Bitter Lesson. We are the first inference-focused frontier AI system, betting early on transformer and transformer-like architectures and on increasing model sizes. Our addressable market is the entirety of inference, unlike many of our competitors.

We are a fully in-person team in San Jose (Santana Row), 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 and work across disciplines as needed.