Welcome to my portfolio! I'm Akash Yadav: A curious AI explorer and MSc Artificial Intelligence student at FAU, Germany. My passion lies in building robust, interpretable, and scalable AI systems from LLMs and GenAI to 3D object segmentation and time-series explainability.With 5 years of mixed experience in software development and AI development, and a strong background in R&D, computer vision, and real-world problem solving, I thrive at the intersection of research and real-world impact.
- 👨🎓 MSc in AI @ Friedrich-Alexander-Universität (FAU), Germany | B.Tech in IT
- 🔬 Current Work: Master Thesis in Uncertainty Estimation on Semantic Segmentation @ ZEISS,
- 🧠 Researcher in LLM Evaluation, XAI, RAG, Vision, and Uncertainty Estimation in Medical Imaging
- ⭐ 1x Hackathon Winner TUM.ai x BKW Engineering
- 💻 Built models for Writer Identification (ICDAR2017), Visual Segmentation, and Ride-Sharing Recommenders
- 📈 Delivered production-ready solutions at Weatherford, WSAudiology, and Amdocs
- 🚀 Achieved F1-score: 0.92+ in real-world ML tasks with customized loss functions
- 📄 Publication in JETIR on attendance monitoring using vision systems
- 🛠 Skilled in developing microservices with Clean Architecture and deploying AI with Docker & Azure
- 🌍 Based in Germany | Open to collaborations in AI/ML research and applied GenAI
- 🏋️ GymLens: XR-RAG application
- 🧠 Time Series XAI: AT-LSTM for Multi-Sensor Fusion
- 📊 Writer Identification & Retrieval using DenseNet (ICDAR2017)
- 🚗 Recommender System for Ride Sharing @ uRyde
- Visual Segmentation of shape by cause
