AI/ML Engineer at Scale AI

Developed production-grade RAG pipelines using Python, LangChain, LangGraph, LlamaIndex, FastAPI, PostgreSQL, and Pinecone, including ingestion, chunking, embeddings, semantic retrieval, and multi-agent orchestration. Built and optimized multi-agent GenAI workflows with planning, retrieval, SQL generation, and tool calling using OpenAI/Claude/Gemini APIs. Engineered cloud-native distributed AI infrastructure on AWS with Docker/Kubernetes and CI/CD tooling. Implemented LLM evaluation and model governance using MLflow and evaluation frameworks, and deployed observability/safety controls with Guardrails, LangSmith, OpenTelemetry, Prometheus, and Grafana. Optimized routing, hybrid retrieval, reranking, and embeddings to support high request volume with low latency.

AI/ML Engineer at Scale AI USA

Developed production-grade RAG pipelines using Python, LangChain, LangGraph, LlamaIndex, FastAPI, PostgreSQL, and Pinecone, including ingestion, chunking, embeddings, semantic retrieval, and multi-agent orchestration. Built and optimized multi-agent workflows using LangGraph and LLM APIs (OpenAI, Claude, Gemini) for planning, retrieval, tool calling, and structured reasoning; improved response quality and reduced hallucinations. Engineered cloud-native AI infrastructure on AWS with containerization and orchestration (EC2, EKS, S3, RDS, Bedrock, Docker, Kubernetes) and implemented automated LLM evaluation and governance via FastAPI/gRPC services, Kafka, MLflow, Ragas/DeepEval, and TruLens. Added AI safety and observability (Guardrails, OpenTelemetry, Prometheus, Grafana) with monitoring, tracing, prompt-injection and PII protection. Optimized hybrid retrieval, reranking, embedding optimization, and agent routing to support high request volume with low latency.