I am an aspiring Data Scientist and Machine Learning Engineer with a passion for turning raw data into actionable insights. Currently, I am seeking full-time opportunities or internships in Data Science, Data Analytics, or Machine Learning.
-
Stevens Institute Of Technology
- Jersey City, New Jersey, USA
-
04:19
(UTC -05:00) - https://rahil-shah-portfolio.vercel.app/
- in/rahil-shah-067b74233
Pinned Loading
-
medical-rag-intelligence-system
medical-rag-intelligence-system PublicA specialized RAG (Retrieval-Augmented Generation) system for healthcare. Uses Gemini 2.0 Flash and FAISS vector search to provide grounded, evidence-based answers with zero-hallucination guardrails.
Jupyter Notebook
-
autonomous-financial-research-agent
autonomous-financial-research-agent PublicAn autonomous multi-agent system built with CrewAI and Gemini 2.0 Flash to automate financial market research. Features specialized agents for real-time data analysis and technical report writing.
Jupyter Notebook
-
multimodal-vision-ai-chat
multimodal-vision-ai-chat PublicA high-performance Multimodal AI Chatbot using Gemini 2.0 Flash to perform visual reasoning, OCR, and object detection. Features robust exponential backoff logic to handle API rate limits and tokenโฆ
Jupyter Notebook 1
-
customer-churn-prediction-api
customer-churn-prediction-api PublicA production-ready machine learning microservice that uses FastAPI and Docker to serve a Random Forest model for predicting customer churn in real-time
Jupyter Notebook
-
urban-traffic-vision-yolov8
urban-traffic-vision-yolov8 PublicReal-time traffic analytics engine using YOLOv8. Detects, tracks, and counts vehicles in city footage to identify peak congestion patterns and optimize urban throughput.
Jupyter Notebook
-
credit-card-fraud-detection
credit-card-fraud-detection PublicEnd-to-end fraud detection pipeline using XGBoost and SMOTE. Achieved an AUPRC of 0.85 by optimizing for precision-recall trade-offs in highly imbalanced financial transaction data.
Jupyter Notebook
If the problem persists, check the GitHub status page or contact support.