Archive
Projects
Project archive
A small pirate-themed GPT trained from scratch on a piratized TinyStories corpus and then SFT-tuned, with a versioned training codebase.
Academic paper discovery tool that profiles a research question, snowballs through citations, filters with an LLM and generates a report.
Benchmark of seven synthetic task families testing whether audio-language models handle trivial questions about event order and timing.
A statically typed, deliberately opinionated toy language with its own parser, standard library and Python interop.
CLIP-based classifier for detecting implicit suggestive imagery, trained on a hand-labelled fashion dataset and published on Hugging Face.
Audio-text retrieval experiments on Clotho comparing a PaSST+RoBERTa baseline against a pseudo-image ViT encoder head.
Fine-tuning run for Falcon H1 7B on tool calling that underperformed the base instruct model, written up in a post.
Browser exercise that walks through Gaussian elimination one row operation at a time.
Web app where JKU students add, share and drill each other's exam questions, with LaTeX rendering, timed tests and an XP indicator.
Journalling app that stores every entry in the user's own private GitHub repo, with Whisper voice-note transcription.
Product discovery engine that scrapes thousands of Shopify storefronts and serves embedding-based search over the aggregate catalogue.
Supervised fine-tuning dataset for tool use, unifying several public tool-calling corpora plus deliberate no-call examples under one schema.
Collection of generative and simulation sketches, particle systems, Mandelbrot set, Maurer roses, self-avoiding walks, cloth waves.
