Dnyanesh Walwadkar

Fine-tuning Large Language Models (LLMs) like GPT-3, BERT, and LLaMA using LoRA/QLoRA. Vision Transformers (ViT) for image classification, object detection, and segmentation. Lightweight model optimization for edge devices and real-time inference. Multi-task and multi-modal learning for facial recognition and 3D reconstruction. Model deployment on resource-constrained hardware with optimized performance. Generative models for synthetic data creation in biometric systems. NLP and Reinforcement Learning for real-world applications. Active participation in Kaggle competitions and open-source contributions.