work Teaching Gemma-3 to Reason via GRPO Three-stage SFT plus GRPO pipeline for reasoning improvement. Agentic Derma Doctor Chatbot Multimodal medical RAG assistant with image and text understanding. Secure Multi-Tenant RAG System Role-aware private RAG for department-specific document access. Diffusion Language Model from Scratch Parallel token unmasking language modeling with diffusion. Smart Product Pricing Ensemble Model Multimodal ensemble for product pricing prediction at scale.
