I analyse measurement data from industrial grids and build machine-learning models for time-series anomaly detection and forecasting. I presented this work at IEEE GPECOM 2026 in Naples. “Machine Learning Power Quality Anomaly Detection as a Foundation for Predictive Maintenance in Industrial Grids” has been accepted for IEEE ENERGYCON 2026 in Paris.
I built an edge device for data acquisition. Firmware in C++ on an ESP32 reads a three-phase industrial power meter over RS-485 (IEC 62056-21 / D0) and a climate sensor over I²C, then publishes JSON over MQTT. I debugged it until it reliably captured BCC-checked OBIS blocks on an 8-second cycle.
I own the Node-RED acquisition flow for the lab test bench. Data from the grid emulator, a Janitza UMG 801, and an OPC-UA load emulator is written automatically to MySQL and InfluxDB, with timestamp checks, UTC handling, and data-quality controls.
I also run an internal documentation portal on a local LLM (Ollama, RAG, FastAPI, PostgreSQL with pgvector), packaged with Docker Compose and used daily by the team.
Since January 2026, and earlier from May to August 2025: I investigate grid instabilities on large MS SQL operating datasets, including unbalanced currents, phase asymmetry, and phase-angle shifts.
I work on a web-based grid simulation platform with integrated ML models in TypeScript and Python, aligned with ISO 50001. I also build Python ETL pipelines and Grafana dashboards that monitor voltage behaviour.
I detected transients and anomalies in voltage waveforms using curve fitting, signal processing, and machine learning. I presented “Robust Transient and Anomaly Detection in Voltage Waveforms Using Curve-Fitted Sinusoidal Models” at IEEE AUPEC 2025 in Brisbane.
Progressed through multiple roles demonstrating leadership and technical excellence. As Software Engineering Manager (August 2025), I guided team members in gathering vital information for web development projects and coordinated cross-functional initiatives.
As Team Lead (July 2025), I led software interns in drone simulation projects using Webots, a sophisticated robotics simulation platform. I mentored the team in implementing Python-based QR code encoding and decoding solutions, and successfully implemented Dijkstra's Algorithm for finding shortest distances in map navigation systems.
As Volunteer AI/ML and Software Engineer (March-July 2025), I learned and implemented advanced software engineering principles, focusing on AI/ML applications and robotic simulations. I gained hands-on experience with Webots for drone virtual simulation, bridging theoretical knowledge with practical implementation in autonomous systems.
Master Thesis Student (September 2024 - February 2025): I served as a Master's Thesis Student on the Integrale Forschung Energie für Elektromobilität: Vom Modell zur Anwendung (iFEMA) project. My research focused on "Analysis and Prediction of Power Quality Anomalies in Power Electronic Dominated Industrial Systems." I employed cross-correlation analysis, curve fitting techniques (sine wave applications), and GridVis software for effective anomaly detection. I successfully identified the most causally significant features contributing to power quality anomalies, which formed the foundation for developing machine learning models capable of forecasting grid quality issues. To address the inherent challenge of imbalanced data in anomaly detection, I implemented specialized imbalanced learning techniques, complementing these with neural network architectures designed for voltage predictions. The technical stack for this work encompassed Python, Node-RED, InfluxDB, TensorFlow, Scikit-Learn, Pandas, Matplotlib, and Seaborn.
Research Assistant (January - August 2024): I developed Python scripts that established reliable communication pathways between Beckhoff PLCs and MQTT servers, creating a robust foundation for industrial IoT applications. I successfully implemented code to publish messages from PLCs to MQTT servers and capture messages from MQTT servers to send to PLCs. I automated critical data logging processes from both PLC systems and MQTT topics into MySQL databases, significantly enhancing data collection efficiency by 40%. My contributions to IoT-based data engineering solutions enabled real-time energy monitoring capabilities. I also successfully transferred variables from MQTT to PLC and back via Node-RED, automated JSON creation for Node-RED, and connected PLCs with Modbus and OPC-UA protocols through Node-RED, leveraging technologies including Python, MQTT, MySQL, Beckhoff TwinCAT (Structured Text), and Node-RED to create an integrated monitoring ecosystem.
Software Engineer (January 2019 - September 2023): I served in the Digital Workplace department, working on the Group Package Factory (GPF) project. I specialized in application packaging using InstallShield and automated project-related tasks through PowerShell scripting, creating automation tools that reduced operational costs by 25% with daily usage exceeding 100+ instances across multiple departments. My responsibilities included working on Azure Intune as a Proof of Concept to streamline application deployment via cloud infrastructure by 30%. I was responsible for providing clients with desired packages according to configurations while maintaining Volkswagen group policies and standards. I automated installation and uninstallation of different applications through PowerShell, thoroughly troubleshooted applications before delivery, and collaborated closely with Quality Analysts as a verification partner to ensure deliverables aligned precisely with customer requirements.
Software Engineer Trainee (September - December 2018): During this formative period, I focused on Robotic Process Automation using Automation Anywhere and successfully automated various project tasks. I gained valuable experience in application packaging through MSI technology. My contributions during this period were recognized with both the Best Debutant Award for exceptional performance in a short timeframe and the Best Innovation Award for creative thinking that reduced manual effort by 20%, demonstrating my commitment to efficiency and innovation within the Volkswagen ecosystem.
Received comprehensive training on the 3dExperience Platform, a cloud-based PLM software. Gained hands-on experience with CATIA including part design, assembly, simulation, and mechanical system behavior analysis. This internship provided foundational knowledge in computer-aided design and product lifecycle management systems.
Gained hands-on experience with Bogie wheel (Tank wheel) testing machines. Designed a robotic arm using AutoCAD and CATIA. Worked in a team of 3 to execute daily activities for testing temperature and failure rates in Bogie wheels, learning valuable lessons in mechanical testing, CAD design, and collaborative engineering work.