Welcome to my GitHub Portfolio! I'm Saravanan M, an IITM Pravartak GUVI Certified Data Scientist with a strong foundation in Machine Learning, Deep Learning, and AI system deployment. I specialize in solving real-world problems using data-driven approaches across healthcare, finance, e-commerce, and NLP domains.
- 🎓 Certification: GUVI - IITM Pravartak Master Data Science Program
- 💻 Background: Passionate Data Scientist with hands-on experience in building and deploying ML/DL models across healthcare, finance, and ecommerce. Skilled in Python, SQL, Streamlit, MLflow, and YOLOv5. Delivered high-accuracy solutions (up to 95%) with a strong focus on EDA, model tuning, and realworld deployment. Proficient in handling structured and unstructured data, and developing end-to-end AI applications.
- 📍 Location: Trichy, Tamil Nadu, India
- 🌐 Portfolio Website: (https://saravananportfolio.com)
I’m passionate about crafting intelligent systems and building AI-powered applications that create measurable impact.
| Category | Tools & Technologies |
|---|---|
| Languages | Python, SQL |
| Data Analysis | Pandas, NumPy, EDA, Data Cleaning, Feature Engineering |
| Machine Learning | Scikit-learn, XGBoost, Random Forest, Logistic Regression, Hyperparameter Tuning |
| Deep Learning | TensorFlow, Keras (DNN, CNN) |
| NLP & GenAI | Text Preprocessing, RAG, LLaMA 2, LangChain, LlamaIndex |
| Visualization | Matplotlib, Seaborn, Plotly, Power BI |
| Deployment | Streamlit, MLflow, Git, Jupyter Notebook |
| Databases & Cloud | MySQL, PostgreSQL, AWS |
| Project | Description | Tools Used |
|---|---|---|
| 🩺 Mental Health Depression Prediction | Predicts depression likelihood based on survey inputs | Keras, Streamlit, Pandas, Matplotlib |
| 🛒 Customer Conversion Analysis | Clickstream-based prediction and clustering to improve e-commerce revenue | Sklearn, Streamlit, Seaborn |
| 🧾 AskDocs AI | Document Q&A system using RAG and LLaMA 2 with GenAI evaluation | LangChain, FAISS, CTransformers |
| 📈 Insurance Cost Prediction | ML-powered app to forecast insurance premium using user profiles | Scikit-learn, Streamlit, MLflow |
| 🫁 Disease Detection with X-rays | YOLOv5-based image analysis to detect chest conditions | YOLOv5, OpenCV, Streamlit |
👉 Explore the Projects section on this repo or my (https://saravananportfolio.com) for demos and source code.
- 📧 Email: [email protected]
- 💼 LinkedIn: linkedin.com/in/itssaravanan
- 💻 GitHub: github.com/Saravanan9698
- 📄 Resume: View Resume
“The best way to predict the future is to create it.”
Let’s collaborate and build impactful, data-driven solutions! 💡