https://github.com/markf94/markf94
Personal Github profile repo.
https://github.com/markf94/markf94
Last synced: 4 months ago
JSON representation
Personal Github profile repo.
- Host: GitHub
- URL: https://github.com/markf94/markf94
- Owner: markf94
- Created: 2021-04-18T18:12:20.000Z (about 5 years ago)
- Default Branch: main
- Last Pushed: 2026-02-27T20:17:32.000Z (4 months ago)
- Last Synced: 2026-02-28T00:37:43.993Z (4 months ago)
- Size: 53.7 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
### Welcome, human! I'm Mark 👋
[](https://uwm.edu/lgbtrc/support/gender-pronouns/)
[](https://maps.app.goo.gl/oJnKkoXgeLYH2bZJ9)
[](https://maps.app.goo.gl/8JfGHrRFA5PS5Vu26)
[](https://goo.gl/maps/oX2GBATc4ev1mgLT6)
I'm a scientific ML leader and entrepreneur with a background in theoretical physics. I co-founded the Toronto-based biotech [ProteinQure](https://proteinqure.com), where I lead the scientific ML team and help build a ML-driven platform for protein and peptide therapeutics (think 3D CAD for molecules). I also co-founded the [Quantum Open Source Foundation](https://qosf.org) (QOSF).
My work focuses on generative models for proteins and peptides, molecular property prediction, multi-property optimization, and scalable production ML systems for drug discovery. At ProteinQure, I work closely with a multidisciplinary team on proprietary ML models for peptide drug delivery and discovery, spanning model development, optimization, experimental validation and real-world experimental integration.
Technically, I spend time on diffusion-based generative models (e.g. [RFDiffusion](https://github.com/RosettaCommons/RFdiffusion), [BindCraft](https://github.com/martinpacesa/BindCraft)) with steering & guidance, multi-objective optimization algorithms such as Monte Carlo Tree Search or evolutionary algorithms (using [deap](https://github.com/DEAP/deap)), Bayesian optimization, and custom ML tooling for scientific workflows. My MSc research explored learning curves of quantum kernel machines, using quantum–classical hybrid models built with built with [PennyLane](https://github.com/PennyLaneAI/pennylane) and [TensorFlow Quantum](https://github.com/tensorflow/quantum).
Quick facts:
- 💬 Ask me about **scientific ML, protein design, drug discovery, Python, quantum computing, and open-source software**
- 📫 Talk to me via *hello@markfingerhuth.ai*
- ⚡ Fun fact: **I like climbing frozen waterfalls in the middle of winter.**
Snapshot:
- 💻 My current favourite open source project is **[Polaris - Benchmarks for AI models in drug discovery](https://github.com/polaris-hub/polaris)**
- 🌱 I’m currently learning: **modular synthesis**, **Vietnamese**, **trad climbing**
- 👯 I’m looking to collaborate on **Protein Drug Discovery** and **Quantum Open Source Software**
- ⌨️ My work setup consists of: [Warp Terminal](https://www.warp.dev/) + [Claude Code](https://code.claude.com/docs/en/overview) + [ralph](https://github.com/snarktank/ralph)
- 🎮 I’m currently playing **ARC Raiders**
- 🚐 I’m currently **climbing frozen waterfalls** and **making music on my modular synthesizer**
[](https://twitter.com/mark_fingerhuth)