B.Tech CSE undergraduate at Rishihood University (Sonipat, Haryana) focused on AI agents, machine learning and full-stack development. Skilled in Python, LangGraph, RAG, LLMs (Groq / Llama), scikit-learn, FastAPI, Streamlit, React and Node.js. Built projects in exam-question difficulty prediction with a RAG agent, infant cry classification from audio, an AI GitHub portfolio reviewer and a full-stack health record manager. Open to remote AI/ML and full-stack internships.
Highlights
- Pro
Pinned Loading
-
Intelligent-Exam-Question-Level-Analysis
Intelligent-Exam-Question-Level-Analysis PublicForked from BuiltByNishant/Intelligent-Exam-Question-Level-Analysis
Predicts exam question difficulty (TF-IDF + Logistic Regression) and writes improvement reports with a LangGraph + ChromaDB RAG agent on Llama 3.3 (Groq). Live Streamlit app.
Jupyter Notebook
-
Infant-State-Recognition-System
Infant-State-Recognition-System PublicClassifies infant cry audio into 8 need states (hungry, tired, belly pain...) using 218 librosa audio features (MFCC, mel, chroma) and a cost-sensitive Random Forest.
Jupyter Notebook
-
student-github-reviewer
student-github-reviewer PublicAI agent that reviews a student's GitHub portfolio: LangGraph pipeline + Llama 3.1 (Groq), FastAPI backend on Render, Streamlit UI.
Python
-
MediTrack
MediTrack PublicPersonal health record manager: upload, organise and share medical reports securely. React + Node/Express + MongoDB with JWT auth.
JavaScript
-
-
Uber_dashboard
Uber_dashboard PublicTableau dashboard analysing Uber trip data: demand peaks by hour, revenue by service category and high-earning route clusters.
If the problem persists, check the GitHub status page or contact support.
