Product & Systems Work
Here you'll find the shipped work: deployed LLM applications, production-ready data systems, and exploratory builds that pushed an idea far enough to learn something concrete.
Each case study explains the problem, the solution architecture, and the lessons that shaped the next iteration — with links to repos or write-ups when you want to dig deeper.
Filter by tooling, topic, or impact to jump straight to the examples that match your roadmap.
A from-scratch implementation of the full transformer → pretraining → SFT → GRPO → GDPO pipeline. Each layer built, tested, and documented. The repo is the artifact, the site is the narrative.
Sequence classification models for personalized size prediction in luxury fashion — LSTMs, attention mechanisms, and a published paper at ACM RecSys 2023.
Local-first RAG pipeline with hybrid search: BM25 + dense retrieval on Elasticsearch, LlamaIndex orchestration, and Llama3 for generation. Evaluated with RAGAS metrics across chunking strategies and retrieval configurations.
Parameter-efficient fine-tuning from first principles — every matrix decomposition derived and implemented in PyTorch without libraries. Validated against Hugging Face PEFT outputs for correctness.
Chat inference on Apple Silicon using MLX — exploring the runtime, quantization options, and packaging story for local LLM deployment with Mistral and Llama2.
No projects matched your filters. Try a different keyword or tag.