Open to opportunities

Heera Dhanvantri

@heeradhanvantri

I build production autonomous AI agent systems and RAG pipelines that reduce LLM hallucinations and automate regulated workflows.

United States

Message

What I'm looking for

I’m looking to build and ship responsible LLM agent products end-to-end—system design through Kubernetes deployment and observability—where measurable ROI, regulated compliance, and reliable, low-hallucination behavior matter.

I’m an AI Engineer with 5+ years designing, building, and deploying autonomous AI agent systems, multi-agent orchestration frameworks, and RAG pipelines in regulated production environments across healthcare and fintech. I’ve delivered $3.3M+ in measurable business value, including 79% process compression in clinical workflows and a 44% reduction in large language model (LLM) hallucination rates.

Most recently as an Agentic AI Engineer, I architected and scaled an autonomous AI agent platform to 500+ daily active users at 99.9% uptime using LangChain, LangGraph, and MCP, while delivering full AI product lifecycle ownership from system design through production deployment. I build reliable RAG and responsible AI guardrails—delivering 91% answer relevance and a 44% reduction in hallucinations—plus production observability and model monitoring with logging pipelines and drift detection. Earlier, I led HIPAA-compliant clinical automation at Centene (prior-authorization compression from 48 hours to under 10 hours) and delivered enterprise automation outcomes at Tata Consultancy Services, including containerized intelligent automation on Azure Kubernetes Service at 99.6% uptime.

Experience

Work history, roles, and key accomplishments

Current

Agentic AI Engineer

Current

Byte Bridge Technologies

Jan 2025 - Present (1 year 4 months)

Architected and scaled an autonomous AI agent platform to 500+ daily active users at 99.9% uptime, orchestrating multi-step workflows with LangChain/LangGraph and MCP to eliminate 30% of manual operations within 90 days. Built RAG pipelines over 1.5M+ documents with Pinecone, achieving 91% answer relevance and a 44% reduction in LLM hallucinations while deploying production systems on AWS ECS with

Lang Chain LangGraph Multi agent orchestration Pinecone Prompt Engineering AWS ECS Docker CI CD

AI Solutions Engineer

Centene Corporation

Jul 2024 - Dec 2024 (5 months)

Led a HIPAA-compliant prior-authorization platform using fine-tuned BERT with LLM decision logic, classifying 78% of requests at 89% accuracy and reducing turnaround time from 48 hours to under 10 hours (79% reduction) with full audit trails. Deployed an LLM member support agent (LangChain + GPT-4) resolving 12K+ daily queries with 87% self-resolution, and built event-driven HIPAA-compliant ETL pi

HIPAA compliance BERT LLMs (LangChain GPT 4 Apache Kafka AWS Lambda Pgvector ETL ELT HL7 FHIR XGBoost

Software Engineer

Tata Consultancy Services (TCS)

Aug 2020 - Jul 2023 (2 years 11 months)

Developed AI-assisted underwriting automation that processed 82% of loan applications 45% faster, and delivered demand forecasting pipelines that cut overstock costs by $800K annually for a $120M+ revenue retail client. Architected and deployed containerized intelligent automation across NLP ticket classification, real-time social sentiment processing (100K+ posts/day), and computer vision defect