Hardware - Thermal Engineer

About FuriosaAI FuriosaAI builds high-performance, high-efficiency AI compute for the Inference Era. Founded in 2017 by veteran semiconductor and AI algorithm engineers, Furiosa operates globally with offices in Korea and Silicon Valley, along with a compiler-focused R&D lab in Lisbon. Our vision is to make AI computing sustainable, enabling access to powerful AI for everyone on Earth. We solve the AI hardware energy and operational cost crisis at the architectural level, rather than through brute force, building the world's first truly AI-native compute platform to unlock the full potential of artificial intelligence for every enterprise. Key Responsibilities • Define end-to-end thermal design across NPU/ASIC package → PCB → card (PCIe FHFL/HHHL, OAM, SXM) → system (server chassis) levels, selecting and designing heatsinks, heat pipes, vapor chambers, cold plates, and TIM • Evaluate trade-offs across cooling approaches — air, liquid, 2-phase, and immersion — and run CFD simulations (FloTHERM, Icepak) including EM-thermal co-simulation using power maps from HFSS/SIwave/PowerDC • Validate thermal performance through wind tunnel testing, hands-on temperature measurement (thermocouples, IR cameras, RTDs), and DOE-based simulation-to-measurement correlation • Support reliability testing including thermal cycling and burn-in • Partner with electrical, mechanical, and manufacturing teams to integrate thermal solutions into packaged hardware, and negotiate thermal specifications with OEM/ODM and hyperscale customers • Author thermal design guides, reference designs, and test reports, and ensure compliance with form-factor standards (OCP OAM, PCIe CEM, NVIDIA SXM, etc.) Minimum Qualifications • Thermal Modeling & Simulation: Utilize computational fluid dynamics (CFD) to simulate heat transfer and fluid flow. • System Design: Design active and passive cooling solutions, including heatsinks, fans, liquid cooling coldplate. • Thermal Testing: Evaluation for NPU card thermal performance with using multi temp recorder and FLIR camera or thermocouples. • Collaboration: Work closely with electrical, mechanical, and manufacturing teams to integrate thermal solutions into packaged hardware. Preferred Qualifications • Experienced GPU or NPU card development or server development for datacenter • Flotherm or Icepak preferred • English communication for business level • Positive and proactive engineer Job Location • Dongtan / Seoul • The primary work location for this position is Dongtan, with the option to work in either Dongtan or Seoul. Candidates who can work in Dongtan are preferred. Why Join FuriosaAI The defining bottleneck of the AI era is building the right hardware and software stack to run it at global scale. Furiosa is solving this challenge holistically from the ground up. With our flagship chip, RNGD, in mass production today and our next-generation platform in development with Broadcom, we are proving that full-stack, tensor-native compute is the future of AI infrastructure. This is a pivotal moment to join our team, right as we accelerate our global expansion. At Furiosa, you will: Solve AI’s Most Urgent Challenge. Help build the high-performance, energy-efficient inference hardware and software required to fulfill the promise of advanced AI. Pioneer Full-Stack Co-Design. Work with teams that are architecting solutions from silicon up through the compiler (featuring innovations like Tensor Contraction Language and Virtual ISA) and serving frameworks. Ship Real-World Silicon, Software, and Solutions. Turn breakthrough technology into commercial deployment. RNGD is in mass production with TSMC and running live enterprise workloads for global leaders like LG AI Research and Samsung SDS. Partner With the Industry's Best. Collaborate across an elite global ecosystem that includes TSMC, Broadcom, SK Hynix, and GUC. Do Your Life’s Best Work. Join a brilliant, low-ego, mission-driven team in a high-trust environment that values autonomy, intellectual curiosity, and shared ambition. Contact • recruit@furiosa.ai