Raghava Alajangi

- Medical image segmentation (semantic and instance) using deep learning methods (UNet, Mask RCNN) - Python packaging, testing, API documentation, and automation using CI/CD. - MLOps: - setting up MLflow experiment tracking server using PostgresSQL and MinIO S3. - data and model versioning with DVC, automated model deployment using CICD. - implemented a GitLab runner on the HPC using Docker, SLURM, and CML. - Developed a web app as a production environment to make models available for end users.

- Developing laser welding image inspection system prototype - Researching data loading techniques - Creating models (transformer-based U-Net, CNN) - Worked on object detection, image classification, and time series analysis tasks - Graph topologies, network analysis, and signal processing - Maintenance of data pipelines, collection, and labeling - Building a web app and ML model integration, containerization (Docker), and deployment

- Deep learning literature review for computer vision tasks - Developing a computer vision application for automatic laser welding image evaluation - Semantic segmentation, landmark detection, image classification - Data pre and post-processing, data labeling, data augmentation - Built models (U-Net, CNN), trained and evaluated - Prediction analysis, hyperparameter tuning, and optimizing models - Developed a web application, integration with ML models, and deployed in production

Projects: Edge AI - Respiratory signal classification with machine learning - Signal processing, feature extraction, statistical analysis, and prepare data - Training and tuning ML models for supervised and unsupervised tasks Edge AI - Anomaly detection in time series data with machine learning - Sensor data acquisition, data engineering, and analysis on the microchip - Train, evaluate, and tuning ML models on an embedded system (AI-chip)

- Logic design and Microprocessors - Advanced control systems - Engineering Mathematics - Physics - Power system simulation with Matlab - Digital electronics - Network theory - Power electronics