ML Engineer, AI Models

Tenstorrent Tenstorrent · Semiconductors · Tokyo, Japan · ML Models

ML Engineer focused on bringing up, validating, and optimizing AI models (LLMs, CNNs, recommendation, vision) on Tenstorrent's hardware and simulators. This role involves porting models into Tenstorrent toolchains, running experiments for accuracy/performance/stability, and debugging cross-stack issues with hardware, compiler, and runtime teams.

What you'd actually do

  1. Bring up and validate AI models such as LLMs, CNNs, recommendation models, and vision models on Tenstorrent hardware and simulators.
  2. Port models into Tenstorrent toolchains and runtime environments.
  3. Run experiments to evaluate model accuracy, performance, and stability.
  4. Debug cross-stack issues and work closely with hardware, compiler, and runtime teams.

Skills

Required

  • Deep learning models in PyTorch, TensorFlow, or JAX
  • Python or C++
  • Neural network architectures, training, and inference workflows
  • Linux
  • Debugging across software, runtime, and hardware
  • Computer Science, Engineering, Applied Mathematics, or equivalent practical experience
  • English proficiency

Nice to have

  • LLM or foundation model inference
  • KV-cache optimization
  • quantization
  • Compiler or runtime engineering for ML workloads
  • Post-silicon validation
  • Board bring-up
  • Firmware development
  • Accelerator platforms
  • Customer or field team experience

What the JD emphasized

  • optimize AI models
  • optimize AI models
  • performance
  • performance

Other signals

  • optimizing AI models on Tenstorrent platforms
  • bring up, validate, and optimize AI models
  • turn research workloads into reliable, high-performance systems