Remote: Yes

Willing to relocate: Yes (Anywhere in the US)

Technologies: Python, SQL, PyTorch, Scikit-learn, Hugging Face, QLoRA, DPO, RAG, LangChain, LangGraph, Ollama, Pandas, NumPy, PostgreSQL, Redis, Pinecone, Qdrant, FAISS, FastAPI, Docker, AWS

Résumé/CV: https://www.sahil-chordia.com/resume.pdf Email: chordiasahil24@gmail.com

MS in Data Science from University of Maryland - College Park (graduated in May 2026, 3.77 GPA). Recently, during internship at Raising the Floor, I built an LLM pipeline that cut metadata audit time by 96 percent and a Qdrant-backed Gemma2 chatbot for assistive technology recommendations.

Recent Projects: 1. Clippy Vision, a local-first AI memory assistant with a 3-stage classification pipeline; 2. Launchway - an AI job application automation platform that got application time under 2 minutes with 86 percent field-fill accuracy. 3. A from-scratch speculative decoding engine studying how DPO alignment affects throughput.

Looking for entry-level AI/ML engineering or backend engineering roles, mainly at funded startups where I can own systems end-to-end.

Currently, its reactive, i.e a simple ReAct loop which calls function to retrieve data from memory and answers question when asked. Everything runs locally (Ollama + a local vision model, SQLite for storage), nothing leaves the machine.

Right now,I am planning to add a skill layer on top of the base watcher. So instead of it just being reactive, it acts proactively, for eg: spotting some mistakes in my code that I may have missed, or when you see X do Y types, and so on.

I am actively building this project. Version 1 is out already. source code: https://github.com/protocorn/clippy-vision