My name is Jia, and I am a Columbia U-trained computer scientist / AI data engineer. I graduated from Williams College with a BA in Applied Mathmatics and Economics. I am interested in NLP, ML, and multimodal LLM with a focus on narrative textual data and semantics processing
-
🔭 I’m currently working on Echo - a language model fine-tuned on chronological textual narratives. What if we train a model that can best predict the direction of its semantic narrative? Aka - can we make a model to predict future events? I write about my process on my substack
-
🌱 Currently working for Faire wholesale in its Strategy & Analytics team
-
🌱 I am also the technical co-founder of Shamiri Health and shipped our first mobile/web-based app as its sole backend engineer; I managed a team of front-end engineers/UX designers.
-
👯 Some other fun apps that I have done
- Power System ROI tool Fun tool that I made for my friend in a climate tech company to quickly qualify leads.
- Recylcomatron Robotics project - made a robot that can use computer vision to sort garbage
- Several language model training projects: Auto captioning where I trained a LSTM model to auto-caption images
- Game Scout Front-end javascript project. A twitter for games
-
Prior - I was an equity derivatives trader at Nomura. I worked on alternative data ingestion and quantitative trading strategies.