Dylan Sandfelder is Co-Founder and CTO of Amorphous AI, an Oxford-based startup building AI-powered solutions that automate the structuring and analysis of unstructured clinical health data in real time.

Dylan earned a BSc with Distinction in Honours Computer Science and a Minor in Mathematics from McGill University, followed by an MSc in Advanced Computer Science from the University of Oxford where his thesis received a Distinction, and is currently pursuing a fully funded DPhil in Engineering Science at Oxford through the Oxford-Man Institute of Quantitative Finance.

Dylan has published peer-reviewed research including a paper on Ego-GNNs presented at IEEE ICASSP 2021, and co-authored work on tactical asset allocation with macroeconomic regime detection published on SSRN in 2025.

Prior to Amorphous AI, Dylan served as a resident applied ML engineer at Kumo, conducted research at the MILA AI Institute in Montreal, and co-founded Piriko, a student community platform that partnered with Concordia University.

Amorphous AI was selected for the Plug and Play UK Health Accelerator Batch 4 Showcase, and Dylan won first place at the Oxford Multi-Agent AI hackathon with an Amorphous AI entry, also securing a Balliol Interdisciplinary Institute grant.