Radhey Mohan

I’m an AI/ML Engineer focused on end-to-end AI systems—deep learning, NLP, and multimodal applications—turning models into production-ready pipelines. I work across LLMs, computer vision, and audio processing, with strong grounding in transfer learning, fine-tuning, and RAG.

In recent work, I’ve built scalable ML pipelines for real-time inference and integrated LLM-based, vision, and audio capabilities using TensorFlow and PyTorch. My approach emphasizes reliability, efficient model architectures, and practical deployment patterns.

I also ship impactful AI projects: an Indian Food Recognition model reaching 89% accuracy with EfficientNetV2 transfer learning, a Smart Document Q&A RAG chatbot using FAISS and LangChain, and a BlogMimic generator fine-tuned with LoRA. I’ve delivered strong outcomes in classification (96% news classifier) and content performance (180+ AI-assisted pieces improving SEO and engagement), guided by prompt engineering and production ML fundamentals.