Description

Machine Learning Israel (MLIL), as part of Annapurna Labs / Amazon, is hiring a Mechanical-Thermal Design Engineer to help define, shape, and integrate the next generation of our ML training and inference accelerators. In this role you will own the mechanical-thermal design of add-in / mezzanine cards, high-power GPU sleds and full ML racks. You will work shoulder-to-shoulder with HW, FW, and silicon teams to drive decisions on cooling solutions, power delivery packaging, structural integrity, and manufacturability, and will lead the mechanical-thermal engagement with our ODM, component vendors and data-center operators.

Key job responsibilities
1. Own the mechanical-thermal design of Annapurna next-generation Machine Learning sled and rack (including thermal solutions and mechanical challenges on PCB / chassis / rack level)
2. Lead and review mechanical and thermal design performed by ODM partners or contractors and component vendors.
3. Debug mechanical and thermal failures on production data center; drive root-cause to closure; take previous learnings to next products
4. Design and execute mechanical and thermal experiments in the lab — from DOE through measurement, analysis, and corrective action.
5. Travel to ODM/manufacturer/vendor sites to monitor and review project kickoffs, device bringups and manufacturing lines