VISWANATH A S
Initializing System...
VISWANATH A S
AI Enthusiast & Developer.
Transforming complex
data into impactful solutions.
Specialized in
Machine Learning, Deep Learning, and Gen AI.
Know More About Viswanath
About Me
Who I Am
I'm an AI enthusiast and developer based in Trivandrum, Kerala. With a foundational background in Physics and Mathematics, I bring a highly analytical, first-principles approach to the world of Artificial Intelligence. I am deeply fascinated by how raw, complex data can be engineered into intelligent systems that solve real-world problems.
What I Do
Data Science professional with practical experience in Python, Machine Learning, Deep Learning, NLP, Computer Vision, and Generative AI. Skilled in developing end-to-end AI workflows including data cleaning, EDA, feature engineering, model building, evaluation, and deployment. Passionate about creating intelligent systems such as predictive models, RAG-based applications, AI automation tools, and data-driven business solutions.
My Mission
Whether I'm developing diagnostic tools for healthcare or engineering highly secure authentication systems, my core goal remains the same: to build scalable, impactful AI applications that push the boundaries of technology. I am constantly learning, experimenting with new models, and looking forward to contributing to innovative, data-driven teams.
Projects
Developed an Ensemble Voting Classifier (Logistic Regression, SVC, Naive Bayes) for disease prediction. Built an NLP pipeline using TF-IDF and text preprocessing. Deployed an interactive Streamlit web app for real-time symptom input and instant disease prediction output.
Built a real-time face authentication system using FaceNet (InceptionResNetV1) and FAISS vector search. Implemented liveness detection using MediaPipe FaceMesh and Eye Aspect Ratio (EAR) algorithm. Designed a data augmentation pipeline with Albumentations.
Created a touchless virtual keyboard utilizing MediaPipe Hand Tracking and OpenCV. Employs advanced hand landmark detection allowing users to type in the air by pinching fingers, with a highly optimized responsive visual interface.
Engineered a Multi-Agent Retrieval-Augmented Generation (RAG) framework using LangChain/LangGraph. Features coordinate agents that perform query decomposition, context retrieval, verification, and final response generation.
Designed an end-to-end NLP pipeline for unsupervised categorization of 1M+ news headlines. Implemented TF-IDF vectorization, K-Means clustering, and Principal Component Analysis (PCA) for high-dimensional cluster visualization.
Skills & Tech Stack
Programming & Databases
Data Analysis
Machine Learning
Deep Learning & CV
NLP & Generative AI
Cloud & Tools
Experience
Data Science Intern
Luminar Technolab, Trivandrum | Sep 2025 – Present
- Developed Machine Learning, NLP, and Deep Learning projects using Python, Scikit-learn, TensorFlow, OpenCV, and YOLO.
- Built and optimized AI models including Regression, Decision Trees, Random Forest, XGBoost, CNN, RNN, and LSTM.
- Implemented Generative AI and RAG applications using LangChain, LLMs, Ollama, Hugging Face, and Prompt Engineering.
- Performed data analysis, visualization, and SQL-based operations using Pandas, NumPy, SQL, and Power BI.
Education
Bachelor of Science in Physics
Govt. KNM Arts and Science College | 2021 – 2024
Minors in Mathematics and Machine Learning (Kerala University)
Higher Secondary Education
St. Joseph’s HSS, Trivandrum | 2019 – 2021
Focus in Computer Science & Mathematics