Principal Research Scientist

PhysicsX is a deep-tech company with roots in numerical physics and Formula One, dedicated to accelerating hardware innovation at the speed of software. We are building an AI-driven simulation software stack for engineering and manufacturing across advanced industries. By enabling high-fidelity, multi-physics simulation through AI inference across the entire engineering lifecycle, PhysicsX unlocks new levels of optimization and automation in design, manufacturing, and operations — empowering engineers to push the boundaries of possibility. Our customers include leading innovators in Aerospace & Defense, Materials, Energy, Semiconductors, and Automotive.

What You Will Do

• Own Research work-streams at a high level to deliver outcomes.
• Align priorities with problem stakeholders, internal and external.
• Set the technical direction for the stream and apply judgement and taste to drive progress.
• Plan roadmaps with clear milestones for key decisions and outcomes.
• Organise and guide the more junior members of the team to effectively execute and deliver against this roadmap.
• Communicate purpose and key outcomes to raise awareness across the company and create opportunities for use and deployment.
• Contribute towards Research group strategy and culture.
• Identify research areas that would be valuable to the company and champion their development, ordering wrt other research objectives.
• Promote effective working patterns and proactively flag issues with team dynamics to foster a productive environment.
• Nurture younger colleagues to grow their skillset and guide their professional development.

The below activities in particular:

• Work closely with our machine learning engineers, simulation engineers, and customers to translate physics and engineering challenges into mathematical problem formulations.
• Build models to predict the behaviour of physical systems using state-of-the-art machine learning and deep learning techniques.
• Discuss the results and implications of your work with colleagues and customers, especially how these results can address real-world problems.
• Collaborate with colleagues beyond the research team to translate your models into production-ready code.
• Communicate your work to others internally and externally as called for in paper publication venues, industry workshops, customer conversations, etc. This will involve writing for academic and non-academic audiences.

What You Bring To The Table

• Ability to scope and effectively deliver projects.
• Enthusiasm about using machine learning, especially deep learning and/or probabilistic methods, for science and engineering.
• Strong problem-solving skills and the ability to analyse issues, identify causes, and recommend solutions quickly.
• Excellent collaboration and communication skills — with teams and customers alike.
• PhD in computer science, machine learning, applied statistics, mathematics, physics, engineering, or a related field, with particular expertise in any of the following:
• operator learning (neural operators), or other probabilistic methods for PDEs;
• geometric deep learning or other 3D computer vision methods for point-cloud or mesh-structured data;
• generative models for geometry and spatiotemporal data (VAEs, Diffusion Models, Bayesian non-parametric, scaling to large datasets, etc.).
• Ideally, >4 years of experience in a data-driven role in a professional industry setting, where you have been instrumental in:
• building machine learning models and pipelines in Python, using common libraries and frameworks (PyTorch / CUDA, ideally with exposure to JAX, NumPy / SciPy), especially including deep learning applications;
• developing models for bespoke problem settings that involve high-dimensional data (spatiotemporal, geometric, physical);
• iterating on network architectures and model structure, tuning and optimising for inductive biases, improved generalisability, and improved performance;
• combining theoretical reasoning with empirical intuition to guide investigation;
• formulating and running experiment pipelines to benchmark models and produce comparable results;
• writing skills for communication of complex technical concepts to peers and non-peers, tailoring the message for the required audience.
• Publication record in reputable venues that demonstrates mastery in your field, and in particular the domains of interest listed above. Desirable venues include (but not limited to): NeurIPS, ICML, ICLR, UAI, AISTATS, AAAI, Siggraph, CVPR or TPAMI/JMLR.

What We Offer

• Build what actually matters – Help shape an AI-native engineering company at a formative stage, tackling problems that genuinely matter for industry and society. This is work with real-world impact.
• Learn alongside exceptional people – Work with a high-caliber, collaborative team of engineers, scientists, and operators who care deeply about doing great work and helping each other improve.
• Influence over hierarchy – We operate with a flat structure: good ideas win, wherever they come from.
• Sustainable pace, long-term ambition – Building meaningful technology is a marathon; we balance focused, ambitious work with a life beyond it. Our hybrid model blends time in the office with work-from-home days.
• And it doesn’t stop there – Equity options, 10% employer pension contribution, free office lunch, enhanced parental leave, nursery scheme, 25 days of annual leave plus public holidays, private medical insurance, Wellhub subscription, eye tests, personal development support, employee assistance programme, bike-to-work scheme and season ticket loan, Octopus EV salary sacrifice.

We are committed to equal employment opportunity regardless of sex, race, religion, ethnicity, nationality, disability, age, sexual orientation or gender identity. We encourage individuals from groups traditionally underrepresented in tech to apply. We sponsor bright women from disadvantaged backgrounds through their university degrees in science and mathematics. We collect diversity and inclusion data solely for monitoring and compliance; data is confidential and used only in aggregate form.