B.Tech undergraduate in Artificial Intelligence and Data Science with hands-on experience across end-to-end machine learning projects spanning deep learning, NLP, and Python-based data analysis. Known for analytical thinking and practical exposure to real-time streams.
Neovent Innovations, Kannur
Completed a 4-week ML internship covering data preprocessing, model development, model evaluation, and practical applications of AI techniques within a structured ML pipeline.
Applied supervised learning workflows on real-world datasets, covering data cleaning, feature engineering, model training, and performance evaluation using Python.
Collaborated with the team on documentation and testing of ML workflows, strengthening proficiency in dataset handling and Python-based predictive analytics.
AI-powered motor health monitoring using LSTM Autoencoder for real-time anomaly detection across 3 multi-sensor time-series streams (MPU6050).
Built real-time dashboards using Flask and Streamlit for reconstruction error visualization, live sensor data tracking, and predictive maintenance alerts.
Integrated Arduino-based LED alert system and tested/validated against live sensor streams for real-time performance monitoring.
AI-assisted decision-support and 3D digital-twin framework (AquaFlow) coordinating water releases across a 4-reservoir cascade to prevent downstream flooding.
Combined multi-horizon LSTM inflow forecasting (1/3/7-day) with Graph Neural Network spatial advisory and Model Predictive Control (MPC) release planning.
Engineered an independent downstream capacity safety layer and deployed a real-time FastAPI/WebSocket backend powering a live 3D Digital Twin and Streamlit operator dashboard.
End-to-end NLP pipeline to classify tweet sentiment across 2 categories (positive/negative), covering text preprocessing, TF-IDF feature extraction, model training, and evaluation.
Implemented SVM-based classification model and deployed an interactive Streamlit web application delivering real-time sentiment predictions on unseen text data.
Template-aware document-intelligence platform that converts CMS-1500 healthcare claim PDFs/images into structured data using PDF rendering, PaddleOCR, and calibrated spatial field extraction.
Built billing-consistency validation (diagnosis codes, NPI, service lines, charge totals) with per-field OCR confidence scoring and SHA-256 deduplication.
Delivered a FastAPI REST backend and browser dashboard for human-in-the-loop review, audit trails, approve/reject workflows, batch processing, and JSON/CSV export.
Vimal Jyothi Engineering College, Chemperi, Kannur, Kerala
Chembilode Higher Secondary School
Chembilode Higher Secondary School
Python for Data Science, AI & Development
Artificial Intelligence Fundamentals — IBM SkillsBuild
GenAI Powered Data Analytics Job Simulation
Solutions Architecture Job Simulation
Machine Learning Internship Certificate