Weights & Biases Expert in Germany

• Developed end-to-end reproducible ML pipelines (PyTorch) with data versioning (DVC), experiment tracking (MLflow), automated testing (PyTest), and CI/CD across all training workflows.
• Scaled Vision Transformer and CNN training across NVIDIA A100 GPU clusters (CUDA, DDP, SLURM); applied hyperparameter optimization (W&B Sweeps) to reduce training overhead and identify optimal configurations.
• Developed a real-time 3D human motion generation system (ViT, VQ-VAE, SMPL-X/PIXIE) for personality-conditioned avatar synthesis; achieved state-of-the-art FID = 6.15 and P-FID = 10.31 on the UDIVA benchmark.
• Validated model expressiveness through structured user studies, achieving 86% accuracy in distinguishing extroverted vs. introverted avatar behaviors.
• Optimized inference pipelines by deploying PyTorch models via TensorRT and ONNX Runtime into native C++ code; benchmarked performance.

• Integration of Post-Quantum Cryptography (PQC) algorithms into high level protocols.
• Security of implementations of Post-Quantum Cryptography algorithms.
• Transition to Post-Quantum public key infrastructures.
• Security evaluations of Post-Quantum Cryptography (PQC) primitives.
• Drone Cybersecurity
• Satellite Cybersecurity
• AI Security

From Crude to Crafted: Refining Participatory Design Data into Stakeholder-Ready Outcomes

• Architected a production Document AI platform using Retrieval Augmented Generation (RAG) over 1,500+ participatory design artefacts to answer historical project queries with grounded responses.
• Designed LLM evaluation combining RAGAS, custom evaluation metrics and human-in-the-loop (HITL) validation workflows to evaluate factual grounding, response quality, and prompt performance.
• Built a React, TypeScript, and D3.js frontend for interactive exploration of AI-generated insights.
• Implemented input layer LLM safety controls and Guardrails, including PII redaction and foul language filtering.

• Benchmarked datasets, collected real-world off-road data (~9000 images) and generated simulation datasets using Unreal Engine.
• Developed Gen-AI image segmentation in Unreal Engine, reducing time from 1-2 days to 5-10 minutes.
• Built Generative AI data enhancement pipeline. Achieved an improvement in synthetic data by +49% mIoU.
• Tech: Python, PyTorch, C++, Git, LangChain, Gen-AI, Linux, W&B, OpenCV, Labelme.

• Designed and implemented multi-modal entertainment toolchains that combine passenger input, vehicle context, large-language models (text-to-text and speech-to-speech) and image generation models to deliver more interactive and immersive in-car experiences.
• Built and orchestrated tools for LLM-based agents, covering session management, background task execution, dynamic user interactions and persistent application state.
• Investigated multi-agent orchestration frameworks for in-car environments, evaluating communication protocols and architectural strategies for coordinated and reliable agent behavior.

• Innovative prompt engineer with expertise in generating and refining prompts specifically for computer science-related images.
• Proficient in developing responses that enhance machine learning models' understanding of visual data in the computer science domain.

• Designed and implemented a real-time perception pipeline using YOLOv7 on Time-of-Flight (ToF) sensor data, enabling live streaming, inference, and on-frame visualization for passenger detection.
• Fine-tuned and evaluated multiple state-of-the-art monocular depth estimation models for Automatic Passenger Counting (APC), and developed a custom hybrid depth model that improved depth accuracy in challenging scene regions.
• Demonstrated that model-generated depth maps outperform raw sensor depth for APC tasks across several datasets, contributing to measurable reductions in counting error.

• Fine-tuned a quantized LLaMA model with LoRA, optimizing hyperparameters for domain-specific, large-scale NLP applications.
• Led development of LLM-based hybrid RAG architectures using the LangChain framework for the legal domain, integrating Document Extraction, Vector Search, Speech-to-Text processing, and Prompt Engineering methods using OpenAI APIs.
• Built an LLM-powered translation service combining OpenAI Whisper for transcription with domain-specific translation and prompting to handle sensitive diplomacy terminology.
• Developed and integrated REST APIs with FastAPI and Pydantic for AI models, collaborating with front-end teams to deploy production-ready applications in secure cloud environments.
• Automated LLM workflows with CI/CD pipelines, containerized models using Docker, and deployed to AWS for scalable cloud infrastructure.

• Focused on exploring the intersection of Music and AI.

• Build data pipelines for raw and curated data layers using AWS S3, Glue, Athena, and Lake Formation
• Establish CI/CD using GitHub Actions or GitLab CI with CodePipeline
• Prototype models into demo APIs packaged with Docker, versioned with Git, added basic tests with pytest, and assist deployments on AWS SageMaker Endpoint
• Perform exploratory data analysis and feature engineering with pandas and PySpark; track experiments in MLflow or Weights and Biases
• Design and execute A/B tests to optimize user engagement and drive data-informed decisions