Mechanical Engineer, Hardware Systems

Normal Computing | Build with Us Normal is an applied AI company solving the hardest problems in AI and silicon. We build foundational hardware and software for the semiconductor industry, critical AI infrastructure, and the broader systems that power our world, in partnership with the world's most advanced institutions. We work as one team across New York City, Silicon Valley (Mountain View), London, Copenhagen, and Seoul.

Normal is building AI inference hardware based on our Thermo In-Memory Computing architecture, designed to deliver orders-of-magnitude more efficient compute. As the founding mechanical engineer on our hardware team, you will define the physical design of that hardware at every level of integration, from the accelerator through the server to the rack. You will help determine what fits, what can be cooled, what can be serviced, and where an existing supplier platform does the job. You report to the head of our hardware systems organization and partner with our silicon, systems, and deployment engineers, and with the external design and manufacturing partners who build to our drawings. We're looking for someone who thrives on turning a system architecture into hardware and takes pride in catching the problem before the build does. The form factor, thermal envelope, mounting, and serviceability decisions you make set the foundation for the systems Normal ships, and the hardware you define is the hardware our customers run. Candidates from OEM and ODM server teams, hyperscaler infrastructure teams, and earlier-stage hardware companies could all be well qualified for this role.

Mechanical Design: Own the electromechanical design of our inference hardware as it moves from engineering configuration to deployable product, including form factor, thermal solution and airflow, connectors and cabling, and the power and thermal envelope it has to live within. System Configuration: Build the mechanical definition of what we ship, from the assembly structure and BOM to the envelope every board and component has to fit within, and identify assembly, cooling, and serviceability problems before parts are ordered. Engineering and go-to-market build against it. Server Integration: Define the mechanical integration of compute, host platform, networking, and power in the chassis, validate airflow and thermals with the system fully populated, and design for serviceability by a technician in a live rack. Rack Integration: Set the rack configuration, power distribution layout, cabling, weight and blanking plan, and the mounting approach across standard and open-rack formats. Build Readiness: Own drawing and BOM releases, manufacturability, supplier engineering reviews, and the resolution of build failures. Weigh the cost and lead-time consequences of design and supplier choices before they are made, and run bring-up and validation on real hardware in our lab. Compliance and Reliability: Own mechanical and thermal qualification, writing the test specifications and procedures, running shock, vibration, and thermal cycling, and taking lab and field failures to root cause. Coordinate safety and EMC certification and environmental compliance with the electrical team and the test lab. What Makes You a Great Fit Six or more years, or equivalent shipped work, owning accelerator or server hardware. You have taken a relevant design from prototype through qualification and production, and can explain what you designed, what failed, and what you changed. Strong mechanical design skills in SolidWorks, Creo, NX, or similar, including FEA, tolerance analysis, and design for manufacturing. You can release drawings that a manufacturing partner can build from. Hands-on thermal design and validation for server or accelerator hardware, with strong first-principles reasoning and practical lab experience. Experience integrating server hardware into standard chassis and racks, with an understanding of mechanical interfaces and