T

Tenstorrent

Designs AI chips and licenses processor cores

Software Engineer AI Model Bring Up

Applied Machine LearningConfirmed live in the last 24 hours
Full-timeTokyo, JapanNo salary listedHybrid

About the Job

Requirements

  • Hands-on experience with deep learning models in at least one major framework such as PyTorch, TensorFlow, or JAX.
  • Strong Python or C++ skills and a solid understanding of neural network architectures, training, and inference fundamentals.
  • Comfort developing in Linux environments and debugging issues across software, runtime, and hardware layers.
  • A degree in Computer Science, Engineering, Applied Mathematics, or a related field, or equivalent practical experience.
  • Business-level Japanese and sufficient English proficiency to work effectively in a global environment.

Responsibilities

  • Port, optimize, and validate AI models such as large language models, convolutional neural networks, recommendation models, and vision models on Tenstorrent hardware.
  • Port and integrate models into Tenstorrent toolchains and runtime environments.
  • Design and run experiments to evaluate model accuracy, performance, and stability.
  • Debug cross-stack issues and collaborate closely with hardware, compiler, and runtime teams.

Desired Qualifications

  • Experience with large language model or foundation model inference, including key-value cache optimization and quantization.
  • Background in compiler or runtime engineering for machine learning workloads.
  • Exposure to post-silicon validation, board bring-up, firmware development, or accelerator platforms.
  • Experience working directly with customers or field teams on AI workload deployment and debugging.
  • Strong spoken English for technical discussions with global hardware, compiler, runtime, and model teams.

About the Company

Tenstorrent

Designs AI chips and licenses processor cores

Tenstorrent designs AI processors and computing systems for developers, enterprises and chip designers that want control over their hardware. Its Galaxy servers connect multiple AI processors to run language models, video generation and other workloads across larger systems. The company sells accelerator cards, workstations and servers, and licenses RISC-V CPU designs and Tensix AI cores for customers to incorporate into their own chips. Its software compilers translate models from popular machine learning frameworks into programs for Tenstorrent hardware. Founded in 2016, it operates from offices in the United States, Canada, Europe and Asia.

Company size
501-1,000
Company stage
Late Stage VC
Total funding
$2.8B
Headquarters
Santa Clara, CA
Founded
2016

Recently Posted Jobs

Explore more opportunities at Tenstorrent

See more jobs