Solution Architect

FuriosaAI is looking for a Solutions Architect to bring the full potential of our powerful RNGD chips/servers to our customers by acting as the primary technical authority in AI/LLM model deployments. From running POCs to benchmarking and debugging, you will translate RNGD’s powerful system to real-world deployments of customers’ models, empowering customers with FuriosaAI’s powerful solutions.

If you are interested in providing the technical expertise in challenging the current status-quo of AI infrastructure in real-world environments, join us in our path to a sustainable future of AI.

• Own end-to-end technical enablement for US customers deploying AI models on FuriosaAI's RNGD NPU using the Furiosa SDK
• Develop POCs, benchmarking studies, and live debugging sessions directly in customer environments
• Act as the technical authority to the US BD/Sales team during pre-sales and enterprise evaluations; translate deep technical capability into business value for engineering and C-suite audiences
• Develop deep, current expertise in FuriosaAI's hardware and software stack and demonstrate it at US technical forums, AI conferences, and customer workshops
• Onboard and train customers on integration patterns, optimization workflows, and best practices post-purchase
• Serve as a technical feedback loop from US customers back to Seoul HQ product and engineering teams

• 2–5 years in a US customer-facing technical role:
  Solutions Architect, Sales Engineer, Forward Deployed Engineer, or equivalent at an AI infra, cloud, or semiconductor company
• Actively current on the AI/LLM landscape — tracking model releases, inference frameworks, and serving stack evolution in real time
• Hands‑on experience with modern inference stacks: vLLM, SGLang, TensorRT‑LLM, Triton Inference Server, or similar
• Hands‑on experience with agent and orchestration frameworks:
  Lang Chain, Llama Index, Lang Graph, Auto Gen, or MCP‑based tooling
• Proficiency in Python; comfortable with DNN frameworks (PyTorch, Tensor Flow)
• Strong written and verbal communication — able to engage credibly with ML engineers at frontier labs and VP/C‑suite executives
• Authorized to work in the US; able to travel to customer sites and to Seoul HQ periodically

• Prior experience at a US AI chip company, cloud silicon team, or AI infrastructure startup
• Familiarity with NPU/GPU accelerator ecosystems, PCIe integration, and data center hardware deployment
• Experience with inference optimization: quantization, kernel tuning, batching strategies, memory bandwidth optimization
• Proficiency in C, C++, or Rust
• Experience working with distributed or cross‑timezone engineering teams