- Built and deployed an end-to-end ML system for automated resume evaluation, processing 300+ CVs/week and generatingstructured features, predictive scores, and actionable feedback, improving resume scores from 5–6 to 9.2–9.6

- Designed a two-layer AI architecture combining GPT-based feature extraction with an XGBoost model trained on 5,000+expert-labelled resumes to score candidates across 14 evaluation dimensions

- Engineered 30+ structured features (experience, skills, ATS keyword density) using prompt versioning and schema validationImplemented LLM feedback and a PostgreSQL feature store for reproducible training and inference

- Developed a production-ready ML pipeline with validation (Pydantic), experiment tracking (MLflow), testing, logging, andFastAPI services for batch and real-time scoring, improving reliability and ensuring consistent deployment

- Managed real-time navigation and risk mitigation in a safety-critical, data-intensive environment.

- Performed quantitative route optimization and critical navigation calculations (ETA, UKC, safety margins)Interpreted multi-source sensor data (ECDIS, radar, AIS, weather systems) to support time-sensitive decision-making

- Applied structured procedures, checklists, and regulatory frameworks to ensure system reliability and compliance

- Collaborated within multinational, shift-based teams with shared accountability for vessel safety and mission execution