I'm a physics PhD student working on [YOUR EXPERIMENT / SUBFIELD — e.g. "precision measurement of ..." or "ultracold atoms"], and on using machine learning to make that experiment work better.
Most of my time goes to [the actual experiment: what you measure, what the apparatus is, what the hard part is]. The rest goes to a question I keep circling back to: which parts of an experiment like this are genuinely bottlenecked on human intuition, and which of those a model could take over.
That’s less glamorous than “AI for science” usually sounds. In practice it looks like [concrete examples — calibration drift, anomaly detection in detector data, learned control of a feedback loop, surrogate models for a slow simulation, automated fitting of messy spectra]. I write most of it up in the blog, including the attempts that didn’t work.
Research interests
- [Your core physics area] — [one line on what draws you to it]
- Machine learning for experimental control — [closed-loop optimisation, learned calibration, whatever applies to your setup]
- [Third interest] — [one line]
Background
- [Year]–present — PhD in Physics, [Your University]. Advisor: [Name].
- [Year]–[Year] — [BSc/MSc], [Institution].
- [Year] — [Internship, visiting position, or notable project], [Place].
Selected work
-
[Paper or project title]
[One sentence on what it does and why it mattered.] arXiv · code
-
[Another one]
[One sentence.] link
Now Currently [what you’re actively working on this month]. Reading [book or paper]. Trying to get better at [skill].
Get in touch
Happy to talk about [topics you actually want emails about] — especially [the specific thing]. The fastest way to reach me is GitHub.