👋 Hey there, I'm Aksel – a 21-year-old programmer with an insatiable curiosity for turning abstract problems into elegant solutions. I'm currently studying Machine Learning & Datascience at Copenhagens University, while working part time at an awesome startup. I love working with everything from machine learning models in Torch, to MPI and BLAS bindings in Rust.
Check out some of my favorite projects:
- Transformer decoder-block implementation (with backprop) in Julia only using standard library.
- Transformer decoder-block reference implementation in PyTorch.
- reverse-mode, auto-differentiation graph in Rust, levereging Generic Associated Types.
- Predicting PDF annotation labels in Julia using KNN and graph embeddings.
- Rust implementation of the Cooley-Tukey FFT algorithm.
- Minimax Chess AI in Rust, complete with a JavaScript front-end and WASM magic. (This was a group project, where I focused on chess logic and the minimax model.)
- Auto-encoder for generating brain images, employing Rust and WASM.
- multithreaded & distributed async runtime in Rust.
I hope you enjoyed learning a bit about me and my projects.
Feel free to connect with me on linkedin or drop me an email if you'd like to get in touch!