Projects
Engineering Intelligence. Delivering Impact.
​
Welcome to my Projects page — a showcase of applied innovation at the crossroads of Artificial Intelligence, Machine Learning, Data Science, and Software Engineering. Each project reflects my passion for building intelligent, data-driven systems that are scalable, explainable, and designed for real-world impact.
From generative AI and predictive modeling to automation frameworks and full-stack applications, my work bridges the gap between research and deployment. With a strong foundation in both ML/DL and data-centric engineering, I focus on creating solutions that are not only technically robust but also production-ready and future-facing.
These projects represent more than just lines of code — they are systems thoughtfully designed with purpose, creativity, and a commitment to solving meaningful, real-world problems.
Generative AI Agents
Designed a Streamlit web application integrating 9 AI agents powered by Google Gemini Generative AI models. The system supports tasks like grammar correction, sentiment analysis, spam detection, invoice Q&A, speech recognition, summarization, and translation, delivering JSON-formatted, explainable results through a unified, scalable, and user-friendly interface.
​
Tech Stack:
Python, Streamlight, Google Gemini API (Generative AI), NLP Techniques, Speech Recognition APIs, JSON, Git
PDF Insight Bot
A Streamlit-based web app built using Retrieval-Augmented Generation (RAG) architecture. It allows users to upload PDFs, ask questions, and receive AI-powered, context-aware answers. The app uses Gemini AI, FAISS, and LangChain for efficient similarity search and text chunking and maintains chat history for seamless interactions.
Tech Stack:
Python, Streamlit, LangChain, FAISS, Google Gemini AI, PyPDF2, GoogleGenerativeAIEmbeddings
Solar Generation Forecating
Developed a deep learning pipeline to forecast hourly solar power generation using multivariate time series data. Implemented LSTM networks to capture temporal patterns. The pipeline includes data preprocessing, sequence generation, model training, validation, and visualization of forecast results.
​
Tech Stack:
Python, TensorFlow/Keras, Pandas, NumPy, Matplotlib
​
​
Parallel Gauss-Seidel Method
Implemented the Gauss-Seidel iterative method for solving linear systems with parallelization techniques. Utilized OpenMP for multi-threading and MPI for distributed computing. Compared execution time and convergence across serial, OpenMP, and MPI versions. Experiments were conducted on the Nova Cluster using the GCC compiler toolchain.
​
Tech Stack: C, OpenMP, MPI, GCC, Nova Cluster
​
​
