About
Harry Bevins
Company Director and Lead Analyst at InferSol Ltd.
I am a consultant specialising in accurate and efficient surrogate modelling of complex simulations and subsequent applications of these models in inference pipelines. I work closely with businesses and researchers to help them unlock insights from their data and improve their operations through advanced analytical techniques.
For me every engagement starts with developing a detailed understanding of the business or research problem, before any modelling. I favour interpretable, well-validated models over black-box complexity, and work closely with clients throughout.
Research Fellow, University of Cambridge
I currently work at the University of Cambridge on inference and surrogate modelling for astrophysics and cosmology.
PhD, University of Cambridge (2023)
During my PhD I developed state-of-the-art machine learning enhanced inference pipelines for studies of the early Universe.
Specialist in Bayesian inference & statistics
With a background in astrophysics, I bring a rigorous, mathematically grounded approach to every data problem.
Our approach
Three principles that guide every engagement.
- 1
Problem first
We start by understanding your problem before reaching for any tools. The right question matters more than a sophisticated answer.
- 2
Interpretable over complex
We favour models you can understand and trust over unecessary complexity. A well-validated simple model is more valuable than an opaque one.
- 3
Close collaboration
We work with you throughout not just at the start and end. You should understand what we're doing and why at every step.
Want to work together?
Drop us a line and we'll get back to you promptly.