Mangesh Gurav

Cloud software for battery (electric) and heat bank (thermal) energy storage. • Battery energy management: price-based scheduling, load and solar forecasting, site monitoring • Heat bank control: MILP charging optimizer on day-ahead prices, heat demand forecasting, MQTT command contract with the on-site controller • Reliability: alarms, deviation tracking, automatic re-planning, device simulator, replay tests on real data

In this role, I had supported research activities in machine learning and artificial intelligence under the guidance of Prof. Juho Mäkiö in the Department of Electrical Engineering and Computer Science.

• Assisted in the migration of the university website from Typo95 to Typo324 (TYPO3 CMS), ensuring accurate transfer of pages, layouts, and content. • Conducted quality assurance by identifying, documenting, and resolving migration issues such as missing elements, formatting errors, and non-functional components. • Collaborated with team members to verify data through Excel sheets and Matomo analytics, cross-checking inputs to minimize errors and ensure content consistency. • Supported system testing and

• Designed and deployed a secure, production-grade MLOps platform on AWS (EKS, S3) using Kubeflow, MLflow, DVC, and Airflow for scalable and compliant ML workflows. • Automated data ingestion, validation, and versioning pipelines for structured, unstructured, and genomic data using Spark, AWS Glue, Great Expectations, and Feast. • Built CI/CD pipelines with GitLab, Jenkins, and Docker/Kubernetes for automated testing, containerization, and model deployment. • Implemented model monitoring, drift detection