Building intelligent systems that bridge data and real-world impact.
I'm an AI Developer with hands-on experience in Machine Learning, Computer Vision, NLP, and LLM fine-tuning. Passionate about programming since the age of 14, with a strong focus on building high-quality, detail-oriented solutions.
Known for a perfection-driven mindset where accuracy, clarity, and performance matter. Proficient in data preprocessing, analysis, and visualization using NumPy, Pandas, Matplotlib, and Seaborn.
Experienced in designing and training deep learning models with PyTorch and TensorFlow, and exploring agentic AI systems for intelligent automation and reasoning-based applications.
Custom regression, classification, and clustering pipelines built for precision and real-world performance. From raw data to deployed model.
CNN-based systems for object detection, face recognition, and image classification. Built with TensorFlow and PyTorch for production-grade accuracy.
Natural language models for sentiment analysis, text classification, and Q&A systems. From simple Logistic Regression to transformer fine-tuning.
Custom fine-tuning of large language models on domain-specific data for specialized tasks — legal QA, education, detection, and more.
Intelligent automation pipelines and reasoning-based agents using the latest techniques in agentic AI and multi-step decision-making.
End-to-end data pipelines with cleaning, feature engineering, visualization, and storytelling using NumPy, Pandas, and Seaborn.
Fine-tuned Legal-BERT model to extract precise insights from legal documents. Built for production-level contract analysis workflows.
CNN-based face detection and identification system using DenseNet121 and TensorFlow. Robust to lighting and pose variation.
LLM-based educational assistant capable of tutoring students across multiple languages. Deployed with Streamlit for real-time interaction.
Predicted eye movements (left, right, blanking) using Digital Signal Processing and a Random Forest Classifier on EOG signals.
Sentiment analysis model classifying reviews as positive/negative. Combined NLP preprocessing with Logistic Regression for clean, interpretable results.
ML-powered phishing URL detector trained to flag malicious websites. Part of the ITSOLERA internship team project suite.
Available for freelance, internships & collaborations.