Nihal Acharya Adde

Data Scientist & Phd Researcher @Volkswagen Group

Wolfsburg, DE

Designed optimization frameworks for hyperparameter tuning and Neural Architecture Search (NAS), improving RL performance by 5% while reducing computational overhead- Worked with my team on vision-language models and diffusion models to simulate edge-case scenarios (e.g, synthetic fog, multi-perspective data, in/out painting), enhancing system robustness, data augmentation, and simulation fidelity- Published research on hyperparameter optimization (LOD 2024 Springer LNCS), NAS for RL (ESANN 2025), HPNAS-RL (DSO@ECMLPKDD 2025)- Mentored two master’s students and an intern, achieving significant contributions to RL optimization projects- Achieved a 5% improvement in RL training performance by deploying lightweight CNN models optimized for real-time applications- Development of simulation environments in Unity3D to generate realistic training scenarios for autonomous driving; connection and control via a Python API for RL model development.