About the Role
The role involves designing and testing novel machine learning algorithms applied to complex physical systems, working closely with researchers to translate theoretical concepts into scalable software solutions.
Responsibilities
- Design and implement machine learning models tailored to scientific datasets
- Collaborate with research scientists to identify modeling opportunities
- Develop software pipelines for training and evaluating models
- Optimize algorithms for performance and scalability
- Conduct experiments to validate model accuracy and robustness
- Analyze results and iterate on model improvements
- Integrate machine learning components into larger simulation frameworks
- Write clean, maintainable, and well-documented code
- Participate in code reviews and technical discussions
- Stay current with advancements in machine learning and computational physics
Nice to Have
- PhD in a relevant field with computational research focus
- Prior work in physics-informed machine learning
- Experience with high-performance computing environments
- Contributions to open-source machine learning projects
- Publications in peer-reviewed conferences or journals
- Knowledge of differential equations and numerical methods
- Experience with containerization and workflow tools
- Familiarity with cloud computing platforms
Compensation
Competitive salary and equity offered based on experience and qualifications.
Work Arrangement
Hybrid work model with a combination of on-site and remote days per week.
Team
Collaborative research environment integrating software engineering, physics, and data science teams.
Research Collaboration
- Work directly with physicists and domain experts to define problems and co-develop solutions
- Engage in interdisciplinary projects that bridge simulation, theory, and data
Technology Stack
- Primary tools include Python, JAX, NumPy, and custom C++ modules
- Infrastructure leverages Kubernetes and cloud-based compute clusters
Sponsorship available for qualified candidates requiring work authorization.
