Open to opportunities

indukuri sainath User

@indukurisainathuser

Message

I build production-grade GenAI, RAG, voice AI, and MLOps systems for real-world products.

India

Message

What I'm looking for

I'm looking to build and deploy production-grade GenAI, RAG, agentic AI, voice AI, and MLOps systems, collaborating with engineering and product teams on scalable real-world AI products.

I'm building production-grade Generative AI systems at Rankinsoft, from AWS serverless RAG applications to real-time voice and chat agents.

For CapitalRAG, I deployed a Bedrock-powered RAG chatbot, built PostgreSQL/pgvector retrieval infrastructure, and improved bulk upserts 63x for 19,000+ records. I also designed seven Prefect pipelines for financial-data ingestion and a 269,649+ record historical backfill.

I architected RANK, a customer-facing AI consultant that supports typed and spoken conversations with preserved context. Its WebRTC architecture delivers sub-second voice latency, while LLM tool calling captures leads in PostgreSQL and syncs them to Zoho CRM.

Previously, I built fraud and risk ML pipelines at Amazon and deployed RAG, semantic search, and LLM inference services at MicroLine Information Systems. I work across Python, FastAPI, Kubernetes, AWS, vector databases, CI/CD, and production MLOps.

Experience

Work history, roles, and key accomplishments

Current

AI/ML Engineer

Current

Rankinsoft

May 2026 - Present (4 months)

Designed and deployed a RAG chatbot on a serverless AWS architecture, integrating AWS Bedrock and OpenAI/Anthropic APIs. Built a vector database layer on RDS PostgreSQL with pgvector and optimized bulk upsert performance by 63x.

Python AWS Lambda API Gateway AWS RDS Pgvector Secrets Manager Cloudwatch Pinecone Prefect Polars Docker MCP

MLOps Engineer

Amazon

Jun 2025 - May 2026 (11 months)

Designed and deployed scalable end-to-end ML pipelines for risk assessment and fraud detection. Built and productionized ML models and implemented CI/CD pipelines using Docker and Jenkins.

Python scikit learn XGBoost CatBoost fastAPI REST APIs Docker Jenkins CI CD BERT SpaCy

MLOps Engineer

MicroLine Information Systems PVT LTD

Jan 2024 - Apr 2025 (1 year 3 months)

Designed and deployed LLM-powered AI systems including RAG-based assistants and semantic search. Built and scaled LLM inference services using FastAPI and vLLM on Kubernetes with GPU support.

Python LlamaIndex Faiss ChromaDB fastAPI VLLM Kubernetes Hugging Face Transformers Azure DevOps Gitlab Docker AWS Azure

AI Consultant

Rankinsoft

Architected RANK, a customer-facing AI agent handling typed and spoken conversation with sub-second voice latency. Built the conversational AI agent on LiveKit Agents SDK and implemented LLM function/tool calling for lead capture.