Software, hardware & machine learning for health.
Researcher and maker, somewhere between code and healthcare. I also like to tinker and build things, from espresso machines to PCBs — sharing what I've been working on.
Projects
all projects →The Nikon Coolscan, Re-implemented
Reverse engineering an aging film scanner's firmware, drivers, and SCSI protocol. Then building a clean-room H8/3003 emulator that NikonScan can't tell apart from the real thing.
Reverse Engineering the Alientek HP20 Reflow Hot Plate
Cracking XOR encryption, decompiling .NET updaters, and reverse engineering BLE protocols on a $100 soldering hot plate.
ChipWhisperer Lite: Self-Assembly with USB-C
Building a ChipWhisperer Lite from scratch in 2025: current BOM, USB-C in place of Mini-USB, Gerbers on GitHub.
Tufte on the Web · A Typographic Swatchbook
A layout system I built for this site, adapted from Tufte's handout style and rewritten for the web with modern touches like responsive margin notes, collapsible sidenotes on mobile, and full-bleed figures.
Research
all research →About
full bio →I'm a final-year DPhil student at the Oxford Big Data Institute. My research applies machine learning to clinical decision-making, with a thesis on antimicrobial stewardship and treatment optimisation in hospital settings.
- research
- Machine learning on electronic health records, focused on antimicrobial stewardship and treatment optimisation in hospital settings.
- before
- Bioinformatics at TUM and LMU München: SARS-CoV-2 drug repurposing, alternative-splicing analysis, and network-constrained biclustering.
- also
- Cooking, building, and breaking things.