I'm currently working on building a start-up, with a pre-seed of $250k from Entrepreneurs First. Before this, I was working on text-to-speech (TTS) as a Machine Learning Engineer at Speechmatics.
I have a Master's in data science from the University of Cambridge. For my thesis (journal publication, Bioinformatics 2025), I created a model for cancer risk prediction that outperforms the best existing approaches in terms of accuracy and interpretability. I developed this work further at Crispin Lab in partnership with Cancer Research UK.
My research interests include AI safety, generative audio modelling, and healthcare AI.
In my spare time, you might find me:
- At the poker table: I was VP of the Cambridge poker team and won Jump Trading's 2024 University Poker Tournament for £12,000. I was also Treasurer of KCL poker soc.
- Staying active: Skiing, bouldering, running, and tennis are a few of my favourites.
- Practicing meditation.
Want to get in touch? Email me at knottenbeltwill@gmail.com. Note: please do not contact me about quant jobs; I'm not interested regardless of compensation.
Publications
CoxKAN: Kolmogorov-Arnold networks for interpretable, high-performance survival analysis
Knottenbelt, W., McGough, W., Wray, R., Zhang, W. Z., Liu, J., Machado, I. P., Gao, Z., & Crispin-Ortuzar, M.
Bioinformatics, 41, Article 8 (2025)