Pravallika

AeroFleet360 is a real-time, AI-powered fleet management platform designed to optimize Boeing’s aircraft operations. Responsibilities included designing and implementing scalable RESTful APIs in Django and FastAPI, developing event-driven microservices with FastAPI + MQTT + WebSockets for real-time alerts, integrating token-based authentication, building predictive maintenance models (scikit-learn, XGBoost, TensorFlow), developing anomaly-detection pipelines (Isolation Forest, One-Class SVM), implementing Remaining Useful Life (RUL) estimation, building interactive dashboards with ReactJS + Redux + Highcharts/Tableau, UI components connected via WebSockets, and deploying ML services on AWS Lambda with Docker. Also involved cloud/resource management (S3, Glue, CloudWatch), automated ML retraining via Apache Airflow, and centralized logging with the ELK stack and CloudWatch monitoring.

Designed and implemented scalable RESTful APIs using Django and FastAPI to expose real-time aircraft telemetry, predictive models, and anomaly detection services. Built event-driven microservices with FastAPI, MQTT, and WebSockets to deliver low-latency alerts and fleet health updates. Implemented token-based authentication and RBAC in alignment with aerospace data policies. Developed ML-enabled predictive maintenance models (RUL estimation, survival analysis) deployed as backend services for real-time inference. Created interactive dashboards with ReactJS/Redux and integrated Highcharts for fleet visualization. Deployed ML services on AWS Lambda and Docker, with automated retraining via Apache Airflow. Implemented centralized logging with ELK and monitoring with CloudWatch.

Developed AeroFleet360, a real-time web application supporting Boeing's fleet management. Designed predictive maintenance machine learning models and built data pipelines for real-time aircraft sensor data analysis. Implemented deep learning for defect detection in aircraft parts and developed dashboards for real-time engine and system insights. Automated model training and deployment with CI/CD and containerized applications for consistent environment management. Integrated solutions across Azure and AWS cloud platforms to monitor and ensure reliability.