Job Description
Tenstorrent is seeking a Software Engineer to join their Product Software Engineering (PSE) team. The successful candidate will be responsible for implementing customer and state-of-the-art ML models on Tenstorrent hardware, ensuring functionality, accuracy, and performance. They will also document and package the implementations for customer use. The role involves working with tt-metalium, tt-nn, tt-forge, and tt-llk software layers, collaborating with software engineers, system engineers, product managers, and customers.
- Implementing ML models using Python and ML frameworks like PyTorch, JAX, ONNX, and TensorFlow.
- Benchmarking, analyzing, and optimizing model performance using C++ and Python.
- Validating the accuracy of implemented models.
- Documenting and packaging models for customer use.
- Collaborating with product managers and other engineers.
- BSc, MSc, PhD or final year of education in Computer Engineering, Computer Science, Software Engineering, Electronics or a related field.
- Experience with implementing ML models on GPUs and other accelerators (advantage).
- Experience with algorithms, data structures, and software development in C++ and Python (plus).
- Familiarity with machine learning, acceleration kernels, ML compilers, high-performance systems, runtime critical systems, and massively parallel processors (plus).
- Competitive compensation package and benefits.