Teams at Tensordyne

Develop agentic workflows and SDK capabilities for bringing up, testing, and optimizing generative AI models on custom accelerator silicon. Focus on LLM inference, numerical optimization, quantization, precision, emulation, model accuracy, latency, throughput, memory efficiency, and distributed execution. Collaborate with model, compiler, runtime, hardware, and architecture teams to deliver reliable software and inform future chip designs.

Develop system-level platform software for high-performance AI inference systems, including firmware, bootloaders, BSPs, device drivers, Linux kernel modules, debugging tools, and tests. Support hardware and silicon bring-up, validation, performance and power optimization, and hardware/software co-design. The role requires close collaboration across software, hardware, silicon, and validation teams, with occasional travel to partners, customers, or manufacturing sites.

Develop system-level software, firmware, tools, and tests for high-performance AI inference platforms. Responsibilities include Linux kernel and driver development, bootloader and BSP work, hardware and silicon bring-up, debugging, profiling, performance and power optimization, and low-level interface integration. The engineer will collaborate with hardware, validation, and silicon teams to deliver production-quality software and support platform deployment in data centers.