I build intelligent Machine Learning systems that move from structured experimentation to scalable deployment.
My work spans modeling, evaluation, and ML infrastructure — with a focus on time-series forecasting, deep learning, and production pipelines.
Currently deepening expertise in attention mechanisms, transformers, and ML system design.
- Machine Learning (model training, evaluation, feature engineering)
- Deep Learning (PyTorch, CNN, LSTM)
- Time-Series Forecasting
- Transformer & LLM Foundations
- ML System Design
- XGBoost
- Matplotlib
- Reproducible training pipelines
- Containerized deployment
- CI/CD for ML workflows
| Project | Focus | Description | Tech |
|---|---|---|---|
| AI Interview Performance Analyzer | Multimodal AI | CNN facial emotion recognition + NLP speech evaluation pipeline with scoring engine | PyTorch, CNN, NLP |
| Car Price Prediction – MLOps Pipeline | Production ML | End-to-end ML with Docker, CI/CD automation, Streamlit deployment | Scikit-learn, Docker, CI/CD |
| Emotion Detection System | Computer Vision | CNN-based facial emotion classification system | PyTorch, OpenCV |
| Depression Detection | NLP | Text-based depression signal detection | Python, Scikit-learn |
- Design and evaluate ML models using structured experimentation
- Compare classical ML and deep learning architectures
- Build reproducible training and inference pipelines
- Develop production-ready ML systems with containerized deployment
- Implement CI/CD workflows for ML codebases
- Translate research ideas into scalable, testable systems
- Attention-based architectures & Transformer fundamentals
- Time-series forecasting systems
- ML system scalability and reproducibility
- Responsible and efficient AI design
🌱 Open to entry-level Machine Learning and MLOps roles