Google Colab Experts in Germany

• LLM Benchmarking: Evaluated and benchmarked 1,000+ LLM-generated responses for technical correctness, code efficiency, safety, and strict adherence to complex technical prompt constraints.
• Mathematical Auditing: Executed highly granular structural audits of model-generated code and multi-step mathematical derivations, documenting root-cause logic gaps, syntax errors, and hallucinations.
• Prompt Architecture: Authored and stress-tested high-complexity mathematical, logical, and coding prompts designed to uncover edge-case failures and assess advanced reasoning capabilities.
• Model Alignment: Led RLHF and SFT workflows, utilizing Google Collab and interactive Python scripts to parse massive datasets of model outputs and deliver highly detailed preference rankings.

• Evaluation and analysis of camera-based traffic light and sign recognition system on various LLM-based autonomous driving systems (LMDrive, BEVDriver)
• Implemented VLM based traffic notice instruction generation unit for closed-loop autonomous driving system which alerts driver in unforeseen driving incidents
• Developed independent LLM-based local chatbot with Llama, DeepSeek and Qwen including MLflow evaluation framework

• 145 hrs.
• Consulting and evaluation for replacing the existing d.velop long-term archive (approx. 36 TB) with a more cost-effective read-only archive solution
• d.velop DMS, Azure, NetApp SnapLock, iTernity iCAS
• Technical and logical consulting on how to set up a future archive system cost-effectively.

• Built a GPT-based synthetic data pipeline that reduced acquisition cost and turnaround time by more than half.
• Modeled audience behavior across underrepresented groups using prompt workflows and statistical validation.
• Evaluated data realism with clustering, regression, and divergence analysis.
• Delivered reproducible Python workflows to automate experimentation in an Agile environment.
• Translated analytical results into clear insights for content and strategy teams.
• Technologies and skills: Python, Generative AI, GPT, Machine Learning, exploratory data analysis, Agile, GitHub, cloud computing, hallucination analysis.

• Published a peer-reviewed article on comparative analysis of biophysical models in diffusion MRI, impacting ongoing research projects.
• Got SciPy selected for the cover image of the corresponding journal issue.

• Help developers with individual live coaching to become pro-level Apache Spark engineers
• Organize and host multi-day, tailored Apache Spark workshops for development teams
• Create educational technical content on a self-hosted blog, YouTube, and social media

• Master's thesis titled "Determining Socioeconomic Resilience to Flood Events Using Machine Learning" as part of the HERAKLION project. Predicted economic damage after floods based on a dataset of 269 samples with 182 features.
• Developed and compared XGBoost, SVR, and KNN using Python, scikit-learn, Pandas, and GeoPandas.
• Achieved a 15–20% improvement in accuracy with XGBoost; evaluated model instability and data distribution effects.
• Identified key data issues like high target variability and weak correlations; investigated the impact of K-Means clustering.

• Led a team of junior Research Engineers, providing mentorship, technical guidance, and career development support to foster their growth in deep learning and machine learning engineering.

• Evaluated 45+ AI use cases and developed a prioritization framework for Siemens’ internal AI roadmap.
• Tested internal tools like DRIM, SiemensGPT, Microsoft Copilot; presented evaluation outcomes to stakeholders.
• Identified performance gaps and improved tool adoption by 65%.
• Supported AI knowledge-sharing initiatives, led tool onboarding sessions, and improved team AI literacy.
• Collaborated with engineering, procurement, and digital teams on tool feedback and strategy alignment.

• Practice-oriented coaching on core aspects of data science
• Teaching advanced concepts in Python as well as automation (with Make and n8n) and ETL processes with Apache Airflow
• Weekly preparation and delivery of practice-oriented programming courses using real-world examples
• Promoting practical programming skills among participants through interactive exercises and individual support
• Developing didactic materials and adapting content to participants' skill levels
• Close collaboration with the team for continuous improvement of course quality and learning outcomes

• Evaluated search results for relevance, accuracy, and quality based on given guidelines.
• Conducted data annotation and content labeling for AI training models.
• Assessed user intent to refine and enhance search engine algorithms.
• Provided linguistic insights for multilingual search optimization.
• Reviewed AI-generated responses to improve natural language processing (NLP).

• Designed, implemented, and evaluated multiple machine learning models (e.g., regression, time series, neural networks) to forecast cryptocurrency prices, incorporating data preprocessing, feature engineering, and model optimization for improved predictive accuracy.
• Performed in-depth data exploration and visualization on large cryptocurrency datasets, using tools like Python and libraries such as Pandas and Matplotlib to identify trends and patterns.

Detecting and classifying vehicles in images and video for traffic monitoring

• A YOLO + Faster R-CNN model built for real-world traffic and autonomous-vehicle scenarios. Awarded Best Paper Award at St Joseph Engineering College, March 2025.

What it does

• The model takes images or video frames and both localizes and classifies vehicles by type, making it usable for downstream applications such as traffic-flow monitoring or perception in autonomous-vehicle systems.

What I did

• Combined YOLO (for fast detection) with Faster R-CNN (for higher-precision classification), rather than relying on a single architecture, trading off speed and accuracy where each mattered most.
• Achieved 90% mean Average Precision (mAP), evaluated using IoU-based metrics rather than just raw accuracy, to properly reflect localization quality.
• Handled the full data processing and evaluation pipeline in Python using TensorFlow and OpenCV.
• The accompanying paper was awarded the Best Paper Award by the Department of CSE at St Joseph Engineering College.

Tech stack: Python, TensorFlow, OpenCV, YOLO, Faster R-CNN