Sathwik Reddy Nethani

I’m an AI/ML Engineer with 5+ years of experience evolving from data analytics to building production-grade Machine Learning, Generative AI, and Agentic AI systems. I focus on secure, scalable solutions that deliver contextual intelligence and reliability in regulated environments.

At Citi Bank, I architected and deployed a secure, enterprise-grade GenAI financial intelligence platform using hybrid RAG (Pinecone + BM25) and Llama 3.1 with Azure OpenAI GPT-4o, improving answer precision by 31% across 750K+ internal regulatory documents. I also designed compliance-aware multi-agent AI workflows with LangGraph and MCP-based tool orchestration, reducing analyst retrieval time by 65% while enforcing RBAC, audit logging, and secure tool orchestration. On top of that, I fine-tuned domain-adapted Transformer models using LoRA/QLoRA in PyTorch, reducing hallucination rates by 28% under internal evaluation benchmarks, and built AI microservices and copilots with FastAPI, Docker, and Kubernetes at p95 latency under 1.8 seconds for 1,500+ users.

Previously, as an ML Engineer at Newgen Software, I delivered production ML for document classification, entity extraction, and intent prediction (improving accuracy by ~27%), and I built scalable Airflow-based pipelines using Python, SQL, and GCP to turn multi-source enterprise data into ML-ready datasets. Before that, at ZeroFox, I processed and analyzed 5M+ daily threat signals, automated ETL workflows, and built anomaly detection models—cutting threat triage time by 30% and improving detection precision by 22%. I’m now seeking to contribute to high-impact AI/ML and Data Science teams building secure, scalable, and business-driven intelligent systems.