SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Gigaton is building autonomous AI control and optimization systems to reduce carbon emissions from heavy industrial manufacturing (cement, steel, glass). The company develops real-time systems that help industrial plants operate more efficiently, stably, and with lower emissions by learning and leveraging manufacturing physics.
As a Research Scientist, you will help shape the research and development agenda by translating challenges from live plants into well-defined research problems. You'll develop and implement methods for modeling, identification, optimization, and control of complex physical systems. Key responsibilities include:
- Designing rigorous experiments, establishing strong baselines, and understanding when and why methods work
- Collaborating with process engineers to incorporate plant physics, operating knowledge, and constraints into models and algorithms
- Working with machine learning and deployment engineers to evaluate promising methods and bring them into production
- Monitoring deployed performance, investigating failures, and using insights to guide further research
- Communicating findings clearly and contributing to the team's scientific understanding
Research areas you may initially work on include: learning dynamics from operational data in noisy, partially observed settings; Bayesian optimization under uncertainty with expensive experiments; dual control and active learning for simultaneous performance improvement and information gathering; and transfer learning across different plants.
The company emphasizes deep tech with real-world impact, working directly with live industrial plants under genuine operational constraints. You'll be part of a team of scientists, engineers, builders, and operators focused on solving hard problems in the physical world.
Benefits include equity (share options), flexible working arrangements, 30 days holiday plus bank holidays, and a generous pension scheme.
REQUIREMENTS:
- PhD in control, machine learning, applied mathematics, engineering, physics, or related field, OR equivalent research experience
- Strong research background in one or more of: control theory, data-driven control, system identification, Bayesian optimization, probabilistic modeling, physics-informed machine learning, imitation learning, or model-based reinforcement learning
- Evidence of research capability through publications, open-source work, or applied research
- Strong Python skills and experience implementing machine learning models using Pandas, NumPy, SciPy, PyTorch
- Interest in working closely with engineers and taking responsibility for real-world method performance
- Practical experience deploying machine learning and control algorithms
- Strong interest in climate mitigation
Note: Domain knowledge of cement/steel/glass manufacturing is not required. Excitement about physics, willingness to learn from operators and process engineers, and curiosity about industrial plant behavior are essential.
