Machine Learning Engineer @ eBay · MS Computer Science, University at Buffalo
I build applied machine learning and NLP systems at eBay. I like turning messy, real-world data into things people actually use — and, off the clock, crafting small, well-made side projects for the people I care about.
- 🔭 Focus: NLP, applied ML at scale, and the data plumbing that makes both work
- 🌱 Always learning: retrieval systems, ML infra, and cleaner ways to ship models
- 💬 Reach me: jayeshscode@gmail.com · jayeshsuryavanshi.com
Machine Learning
Engineering
- Social-Unrest-Prediction-NLP — transformer pipeline predicting social unrest from ACLED data (extraction → summarization → event prediction)
- Sentiment-Analysis-IMDb-Reviews — clean PyTorch RNN / bi-RNN / bi-LSTM sentiment classifiers, compared head to head
- POS-Tagger-for-Hindi-Language — low-resource NLP: an n-gram back-off POS tagger for Hindi (~88% accuracy)
- Lyft-Motion-Prediction — a two-level CNN → transformer motion-prediction pipeline for the Lyft Kaggle challenge
- Walleto — full-stack e-wallet (Angular + Spring Boot / MySQL) for a bit of SWE range
Off the clock: chasing national parks 🏔️, overengineering my personal finances, and curating far too many Spotify playlists 🎧