C++ ML Engineer — High-Performance Model Training

Tenstorrent Inc.; Gdańsk, Warszawa; Hybrid; PLN 180,000 - 320,000; Full time

Job summary

Tenstorrent Inc. is seeking a C++ Machine Learning Engineer to work on the training framework behind our advanced models. The role focuses on high-performance C++ code, and shaping how new layers and operators are implemented to scale models on our custom silicon.

Based in Poland with a hybrid setup in Warsaw or Gdańsk (remote within Poland possible). We welcome candidates across experience levels and align offers accordingly.

Qualifications

• Strong in C++ and low-level systems programming with performance in mind.
• Familiar with tensors, memory layout, and compiler graphs.
• Experience with PyTorch and how frameworks map to hardware.

Responsibilities

• Extend and optimize the ML training framework with new ops and layers.
• Debug and tune model performance on Tenstorrent chips.
• Collaborate with compiler and kernel teams to ensure models compile and run as expected.
• Support integration of real-world models into production.

Skills

C++

Low-level systems

Tensors

Compiler graphs

PyTorch

Tools

PyTorch

Job description

Tenstorrent Inc. is seeking a C++ Machine Learning Engineer to work on the training framework behind our advanced models. The role focuses on high-performance C++ code, and shaping how new layers and operators are implemented to scale models on our custom silicon.

Based in Poland with a hybrid setup in Warsaw or Gdańsk (remote within Poland possible). We welcome candidates across experience levels and align offers accordingly.