Andrew Friedman

afriedman412 [at] gmail [dot] com

Home • Work • Projects • Open Source • Writing • Content

ABOUT

My name is Andrew, but most people call me Andy.

I am an engineer, data scientist, coder, writer and talker. I wear a lot of hats and like to learn through doing.

Recently completed my M.S. in Data Science at Northwestern.

I believe the best data is rare data, and I am particularly interested in data acquisition from difficult sources.

I used to write for a living, and I still write a lot. I also used to DJ a lot. I am still technically signed to Tommy Boy Records. Before that, I did genomic research at NYU.

SKILLS

BROADLY:
NLP and LLM systems, RAG pipelines, structured extraction, model evaluation, fine-tuning and domain adaptation, embedding-based retrieval, production ML platforms, backend engineering, data pipelines.

TECHNICALLY:
Python, SQL, REST APIs, microservices, Docker, GitHub Actions / OIDC, CI/CD, unit testing, GCP (Cloud Run, Cloud Scheduler, Cloud SQL), AWS, Linux, GPU-backed inference, containerized workloads.

SPECIFICALLY:
PyTorch, LightGBM, Hugging Face, LoRA fine-tuning, FAISS, ChromaDB, sentence-transformers, OpenAI / Anthropic / Ollama APIs, FastAPI, Flask, Pandas, NumPy, Scikit-learn, SpaCy, Plotly, BeautifulSoup, Selenium.