Skip to content
View JayeshSuryavanshi's full-sized avatar

Block or report JayeshSuryavanshi

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
JayeshSuryavanshi/README.md

Hi, I'm Jayesh 👋

ML Engineer @ eBay · NLP & applied ML at scale


I build applied ML and NLP systems from messy, real-world data — and small, well-made side projects off the clock.

Now · retrieval systems, ML infra, and cleaner ways to ship models.

Selected work

  • Social-Unrest-Prediction-NLP — transformer pipeline forecasting social unrest from ACLED data (extraction → summarization → event prediction)
  • Sentiment-Analysis-IMDb-Reviews — PyTorch RNN / bi-RNN / bi-LSTM sentiment classifiers, compared head to head
  • GeoSocialX — geospatial toolkit for geotagged social data (CSV · GeoJSON · Bluesky), with mapping built in

Website · LinkedIn · Email

Pinned Loading

  1. social-unrest-forecasting social-unrest-forecasting Public

    2026 revival of a 2022 NLP course project: honest out-of-time forecasting of social unrest on ACLED/GDELT. Finding (adversarially reviewed): news text adds no skill beyond the event record.

    Python

  2. GeoSocialX GeoSocialX Public

    Map the geography of X posts, fetch geotagged posts by radius, then find hotspots and time trends. Pure-stdlib analysis, optional interactive maps.

    Python 1

  3. category_encoders category_encoders Public

    Forked from scikit-learn-contrib/category_encoders

    A library of sklearn compatible categorical variable encoders

    Python

  4. DeepOD DeepOD Public

    Forked from xuhongzuo/DeepOD

    Deep learning-based outlier/anomaly detection

    Python

  5. l5-motion-prediction-study l5-motion-prediction-study Public

    Motion prediction on the Lyft Level 5 dataset under a single-laptop compute budget

    Python

  6. yzhao062/pyod yzhao062/pyod Public

    A Python library for anomaly detection across tabular, time series, graph, text, image, and audio data. 60+ detectors, benchmark-backed ADEngine orchestration, and an agentic workflow for AI agents.

    Python 10k 1.5k