Foundation models for scientific and industrial data.
Expert AI: foundation models for the data that runs science and industry
We define Expert AI as foundation models built for scientific and industrial data: continuous and irregular time signals, spatiotemporal fields, multimodal scientific recordings, time-stamped event logs from real organizations. The kind of data that doesn’t look like text, doesn’t fit transformer assumptions, and rarely leaves the institutions that hold it. This isn’t a fine-tuning problem. It requires architectures designed for the data, infrastructure designed for how the data lives, and a tech stack of AI components purpose-built to assemble foundation models for new domains.
General-purpose AI was never built for science and industry. The data is sparse, noisy, distributed, and governed by strict rules. It demands expert intelligence: models that understand the structure of a domain, not just the surface of its data.
Our approach is co-foundation with domain experts: AI and domain science evolving together. We build domain-specialised foundation models trained to master a field’s data, constraints, and logic from the ground up. We work hand in hand with the scientists, clinicians, and engineers who know the domain best. Our models learn from distributed sources through decentralised architectures, keeping sensitive data where it belongs.
We anchor in brain and neuroimaging, the most complex signal of all and the hardest class of expert data we know how to characterize. In parallel, our cross-domain transfer programme works on extending the same components and methods to open new application domains.
We invest in demonstrators that turn research into tangible proof, working with partners on concrete use cases that test our models in real conditions and inform the next iteration of the science.
Our team finished 2nd out of 1,183 in the NeurIPS 2025 EEG Challenge, and we publish at Tier-1 AI conferencesincluding NeurIPS, ICML, and CVPR. This dual focus, rigorous science and applied impact, defines our approach.
The role
We’re hiring a Research Scientist (Reinforcement Learning) to help bring RL into the core of how foundation models are adapted and improved for industrial use.
In industry, models don’t live in isolation: domain experts validate, correct, and act on model outputs. We want RL to leverage that expert feedback not only as a post‑training patch, but increasingly earlier in the pipeline, shaping objectives, training signals, and adaptation strategies.
What you’ll work on
Human‑in‑the‑loop reinforcement learning
Turn expert validation/correction into a reliable learning signal.
Design feedback interfaces/signals that are practical in real operational settings.
RL for industrial foundation models
Develop RL methods that sit on top of (or integrate with) foundation models used in production.
Explore ways for RL to intervene earlier in the chain (not just after deployment).
From research to deployment
Build evaluation protocols aligned with real constraints: robustness, uncertainty reduction, safety, auditability, and cost of error.
Work closely with scientists/engineers to ship demonstrators that connect benchmarks to field outcomes.
What we’re looking for
PhD (preferred) or equivalent research experience in Reinforcement Learning / Machine Learning.
Strong foundations in RL (e.g., policy optimization, off‑policy learning, offline RL, exploration, credit assignment).
Ability to design rigorous experiments, debug failure modes, and iterate fast with scientific discipline.
Strong programming skills (Python; deep learning stack such as PyTorch).
Nice to have
Experience with real‑world RL constraints (noisy/limited feedback, safety requirements, deployment considerations).
Comfort with complex data modalities (time series, scientific/industrial signals, multimodal setups).
Publications or open research artifacts in RL / sequential decision‑making.
Why join
Work on RL problems that matter in the real world: expert feedback loops, uncertainty reduction, and mission‑critical constraints.
A research culture that values clarity, rigor, and humility and that connects fundamental ideas to deployable systems.
High ownership in a small team: you’ll shape direction, not just execute tasks.
Recruitment prescreen (30-45min)
Scientific deep dive (remote-45min)
Half day of scientific interview (Architecture - Coding - Research talk) + Culture fit
References call