Suraj Bhardwaj
AI Engineer & Researcher | ML/DL, GenAI, Agentic AI
AI Engineer and Researcher
I am an AI Engineer working across computer vision, natural language processing, and multimodal AI. My expertise spans from developing label-free algorithms like the Clustered Feature Weighting (CFW) method—which addressed long-tailed data distributions and improved cross-modality adaptation in large driver distraction detection datasets-to building retrieval-augmented generation (RAG) systems and multi-agent chatbots. I hold an M.Sc. in Mechatronics with a specialization in Artificial Intelligence from Universität Siegen, where my thesis on self-supervised driver distraction detection was conducted in collaboration with Fraunhofer IOSB and Compuet Vision Group Uni-Siegen under the guidance of Prof. Michael Möller, Dr. Jovita Lukasik and M.Sc. David Lerch.
At Fraunhofer IOSB, I contributed to Human–AI Interaction projects including KARLI and SALSA, working with Dr.-Ing. Michael Voit, Dr.-Ing. Frederik Diederichs, and M.Sc. David Lerch. There, I integrated vision–language models and multimodal pipelines for driver attention analysis and sleep stage recognition. Earlier, in the Visual Computing Group led by Prof. Margret Keuper, I worked on Out-Of-Distribution Robustness, GANs, Latent Diffusion Models, and Neural Radiance Fields (NeRFs), building strong foundations in computer vision and generative artificial intelligence.
news
| Nov 18, 2025 | Project Alert: Developed the GDPR RAG Assistant – Evaluation-First Legal Compliance Chatbot. Framed and implemented the problem of trustworthy GDPR Q&A: a RAG system that provides auditable, citation-backed answers and clearly signals when the knowledge base lacks coverage. |
|---|---|
| Jul 07, 2025 | Publication Alert: My paper “Self-supervised Driver Distraction Detection for Imbalanced Datasets” got accepted for publication and presentation as full paper in the IEEE 28th International Conference on Intelligent Transportation Systems (ITSC 2025). |
| Sep 19, 2024 | KARLI Final Event: Led technical demonstrations of a Level 3 Mercedes-Benz Advanced Occupant Monitoring System, communicating its machine learning pipeline and real-world relevance to investors, scientists, and public sector officials. |
selected publications
-
Self-supervised Driver Distraction Detection for Imbalanced DatasetsIn Proceedings of the IEEE 28th International Conference on Intelligent Transportation Systems (ITSC), Nov 2025