Machine Learning Software Engineer, Research

PhysicsX is a deep-tech company focused on accelerating hardware innovation through AI-driven simulation software. The Machine Learning Software Engineer will work closely with research scientists to design and optimize machine learning models for real-world physics and engineering challenges, contributing to the development of high-fidelity simulations across various industries.

Responsibilities:

• Work closely with our research scientists and simulation engineers to build and deliver models that address real-world physics and engineering problems
• Design, build and optimise machine learning models with a focus on scalability and efficiency in our application domain
• Transform prototype model implementations to robust and optimised implementations
• Implement distributed training architectures (e.g., data parallelism, parameter server, etc.) for multi-node/multi-GPU training and explore federated learning capacity using cloud (e.g., AWS, Azure, GCP) and on-premise services
• Work with research scientists to design, build and scale foundation models for science and engineering; helping to scale and optimise model training to large data and multi-GPU cloud compute
• Identify the best libraries, frameworks and tools for our modelling efforts to set us up for success
• Own Research work-streams at different levels, depending on seniority
• Discuss the results and implications of your work with colleagues and customers, especially how these results can address real-world problems
• Work at the intersection of data science and software engineering to translate the results of our Research into re-usable libraries, tooling and products
• Foster a nurturing environment for colleagues with less experience in ML / Engineering for them to grow and you to mentor