Skip to content
View IshaanPotle's full-sized avatar
👋
👋
  • Quantum Pulse Consulting
  • Stony Brook, New York
  • 08:30 (UTC -04:00)
  • LinkedIn in/ishaan-potle

Highlights

  • Pro

Block or report IshaanPotle

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.

Content in all repositories owned by your account will be closed.
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
IshaanPotle/README.md
Ishaan Potle — Data Science · Applied AI · Graph Learning

About

name:      Ishaan Potle
role:      M.S. Data Science @ Stony Brook University
focus:     [ Applied AI, Health & Risk Modeling, Graph Learning ]
exploring: [ Graph Neural Networks, Distributed Systems, System Design ]
open_to:   [ Research collabs, OSS contributions, Full-time roles ]

Tech Stack

LANGUAGES

Python, R, C++, TypeScript, Bash
ML & DATA

PyTorch, TensorFlow, scikit-learn, OpenCV, Anaconda
FRAMEWORKS

FastAPI, Flask, React, D3, Next.js
DATA INFRA

Postgres, Redis, Kafka, Elasticsearch, MongoDB
CLOUD & DEVOPS

Docker, Kubernetes, AWS, GCP, Grafana, Prometheus
TOOLING

Git, GitHub, Linux, LaTeX, Notion

Featured Projects

Privacy-preserving federated learning pipeline for clinical risk scoring — trains across sites without centralising patient data.

Graph neural network that flags money-laundering rings by learning over transaction graphs instead of isolated rows.

Retrieval-augmented clinical assistant with grounded citations and guardrails against unsupported medical claims.

Fairness-aware ranking system that audits and mitigates demographic bias in automated candidate screening.

Network analysis of global food trade to quantify shock propagation and surface fragile supply routes.

Running notebook lab — modelling ideas, benchmarks and write-ups that later graduate into full projects.

Stats

GitHub stats Contribution streak

Top languages

Contribution snake animation

3D contribution calendar

LeetCode stats

Thanks for visiting

Pinned Loading

  1. Bias-Resistant-Resume-Ranker Bias-Resistant-Resume-Ranker Public

    Bias-Resistant Resume Ranker uses Python, FastAPI, and advanced NLP (GPT-based) models to score and rank resumes while detecting and mitigating bias. It features modular microservices, custom fairn…

    Python

  2. Data-Science-Sandbox Data-Science-Sandbox Public

    Full-stack ML platform with React (TypeScript) frontend and FastAPI backend. Supports end-to-end workflows: data wrangling, manipulation, feature engineering, selection, scaling, modeling, hyperpar…

    Python

  3. Federated-Health-Risk-Prediction-System Federated-Health-Risk-Prediction-System Public

    Federated Health Risk Prediction System: A production-ready platform using Flower for federated LSTM training on FHIR/Fitabase data, with SHAP explainability, robust data pipelines, and a Streamlit…

    Python

  4. Global-Food-Supply-Chain-Resilience-Analyzer Global-Food-Supply-Chain-Resilience-Analyzer Public

    Predicts global food supply chain disruptions by processing 300k+ climate, economic, and geopolitical data points via an Airflow ETL pipeline. Utilizes LSTM (TensorFlow) and ARIMA (Statsmodels) mod…

    Python

  5. Graph-Neural-Network-AML-Detection-Engine Graph-Neural-Network-AML-Detection-Engine Public

    Production-ready AML detection system using Graph Neural Networks (GraphSAGE, GAT, TGN) for real-time fraud detection on crypto transaction graphs. FastAPI backend, Streamlit dashboard, Dockerized …

    Python 1 1

  6. LLM_Medical-Chatbot LLM_Medical-Chatbot Public

    A retrieval-augmented medical chatbot combining LLMs with the MedQuAD dataset. User queries are semantically matched to trusted medical Q&A using vector search, then contextualized for the LLM to g…

    Python