Erik An
Mathematics and Data Science, University of Vienna
Software engineering background, moving through mathematics toward computational neuroscience and brain-computer interfaces.
ongoing
-
mfds ·
My entire bachelor's degree done in plain text and run as a software project. Emacs and org-mode throughout, python and julia for tooling and coursework. Lecture notes, problem sets and exam preparation in latex. Math notation standardized. The directory layout is a written spec with a validator, run as a pre-commit hook. Sources and generated products are separated. LLMs through the API handle extraction and aggregation at volume, not mathematics.
selected work
-
total_perspective_vortex ·
Motor-imagery classification over PhysioNet's EEGMMIDB. Common Spatial Patterns written from scratch as a scikit-learn transformer, feeding LDA; 0.63 mean held-out accuracy over 109 subjects and 6 binary task groups. All derivations (CSP + architecture) sit beside the implementation in docs/; CSP is checked against mne.decoding.CSP.
-
audio_print ·
A Shazam-style audio identification engine: spectrogram analysis, spectral peak extraction, and combinatorial hashing matched against a SQLite fingerprint database. 97.8% identification on 138 noisy clips against an 8,000-song database, ~260 ms per query.
-
transcendence ·
Browser Pong with accounts, 2FA, tournament system, three microservices behind a WAF, TLS and secrets issued by Vault. My part was the game (with an AI opponent that sees the board once per second and predicts where the ball lands), most of the frontend and half of the match service. Canvas engine from scratch, SPA hand-written in TS, no frontend framework. Final project of the 42 Common Core.