Abinav Prasad

Background

About Abinav Prasad

I build production AI systems that serve real users at scale.Close to 3 years of experience, I've architected and deployed ML solutions across conversational AI, computer vision, and recommendation systems – serving 100K+ daily users with measurable business impact.Key projects- Conversational AI platform: LangGraph + RAG over 200K+ embeddings, 85% intent accuracy, 35% faster purchase funnel- Visual search: Custom Siamese networks, 87% accuracy, PyTorch production deployment- Learn-to-Rank: XGBoost algorithm, 30% CTR increase through A/B testing- Hybrid recommendations: Sub-100ms latency, 5-10% conversion liftTech stack: PyTorch, TensorFlow, LangChain, LangGraph, FastAPI, OpenSearch, FAISS, AWS, DockerI focus on what happens after the model trains: production deployment, experimentation, monitoring, and actual business impact.Currently exploring how LLMs and RAG architectures apply beyond domains