Ethos

We aim to do excellent science in an atmosphere that is serious about the work and relaxed about everything else. Good research takes hard work, patience, technical depth, and honesty about what the evidence does and does not show.

How we work

Our work is curiosity-driven. We follow problems wherever they lead, and return to questions we thought we understood with new methods, new data, and a sharper sense of what the answer should look like. T. S. Eliot put it better than we could:

The group runs on regular group meetings, individual meetings, and informal discussion. Everyone should have enough structure to make progress and enough space to think independently. We value constructive criticism: direct enough to improve the science, careful enough to keep the discussion useful.

We pair cutting-edge science with cutting-edge tools. Our work spans DNS and LES, theory, data assimilation, and machine-learning surrogates, and we adopt new methods quickly when they earn their place — including agentic engineering for software, data, and analysis workflows. The bar is the same in every case: methods, data, and interpretation that can stand up to close scrutiny.

What we value

• Excellence — methods, data, and interpretation that can stand up to close scrutiny, and a track record of doing things well.
• Hard work — steady, reliable effort over long periods.
• Honesty — clear reporting of assumptions, uncertainty, mistakes, and limits.
• Constructive criticism — rigorous feedback aimed at making the work better.
• Inclusivity — a group culture where people can contribute fully and be taken seriously.
• A sense of humour — research is long and the failure modes are inventive. Laughing at a stubborn bug, a misbehaving solver, or a reviewer's odder remarks is much healthier than taking any of it personally. Take the work seriously; don't take yourself too seriously.

Mentoring

Mentoring is central to how the lab works, not an optional extra. Students and postdocs should leave the group as stronger scientists, clearer writers, better programmers, and more independent researchers than when they arrived — and helping that happen is part of everyone's job, including mine.

Mentoring is not only top-down. We don't all have the same skillsets: one person's deep knowledge of turbulence theory sits next to another's fluency with HPC, a third's intuition for experiments, a fourth's command of modern software tooling. Peer mentoring — explaining what you know, pairing on a hard problem, reviewing each other's code or drafts — is often the fastest way for everyone in the room to get better, the mentor included. Take the time to teach, and take the time to learn from the people around you.