AI Engineer with 3+ years of experience designing and deploying production-grade agentic AI systems, RAG pipelines, and multi-agent orchestration workflows using LangGraph, LangChain, and CrewAI. Shipped LLM applications at enterprise scale (Humana), including fine-tuned NLP/BERT models, vector retrieval architectures, and evaluation/observability pipelines to improve reliability in regulated, HIPAA-compliant environments. Hands-on with tool/function calling, human-in-the-loop guardrails, and cost-aware model routing across AWS and Azure. Open-source contributor who built Nectr, an agentic PR-review system with Neo4j dependency graph mapping and persistent cross-session memory (Mem0).…

Dhanush Chalicheemala

AI Engineer with 3+ years of experience designing and deploying production-grade agentic AI systems, RAG pipelines, and multi-agent orchestration workflows using LangGraph, LangChain, and CrewAI. Shipped LLM applications at enterprise scale (Humana), including fine-tuned NLP/BERT models, vector retrieval architectures, and evaluation/observability pipelines to improve reliability in regulated, HIPAA-compliant environments. Hands-on with tool/function calling, human-in-the-loop guardrails, and cost-aware model routing across AWS and Azure. Open-source contributor who built Nectr, an agentic PR-review system with Neo4j dependency graph mapping and persistent cross-session memory (Mem0).…

Available to hire

AI Engineer with 3+ years of experience designing and deploying production-grade agentic AI systems, RAG pipelines, and multi-agent orchestration workflows using LangGraph, LangChain, and CrewAI. Shipped LLM applications at enterprise scale (Humana), including fine-tuned NLP/BERT models, vector retrieval architectures, and evaluation/observability pipelines to improve reliability in regulated, HIPAA-compliant environments.

Hands-on with tool/function calling, human-in-the-loop guardrails, and cost-aware model routing across AWS and Azure. Open-source contributor who built Nectr, an agentic PR-review system with Neo4j dependency graph mapping and persistent cross-session memory (Mem0).

See more

Work Experience

AI Engineer at Humana
July 1, 2025 - Present
Built a multi-agent prior authorization pipeline using LangGraph with dedicated stages for clinical NLP extraction, policy retrieval, and medical necessity matching. Added human-in-the-loop interrupt nodes for denial routing, reducing PA decision time by 40%. Developed a RAG pipeline over Medicare Advantage policy documents using LangChain and ChromaDB on Azure Databricks, applying semantic and hybrid retrieval to match submitted clinical data against payer coverage criteria with 91% policy alignment accuracy. Fine-tuned BERT NER models on unstructured provider submissions to extract CPT codes, ICD-10 diagnoses, and patient history, routing structured outputs into a downstream PA matching engine and reducing manual clinical review by 30%. Instrumented the pipeline with LangSmith tracing and RAGAS evaluation for hallucination, retrieval faithfulness, and node-level latency monitoring, incorporating reviewer annotations into a HIPAA-compliant self-improving evaluation loop. Collaborated
Machine Learning Engineer at CitiusTech
July 1, 2022 - August 1, 2024
Built an XGBoost + SMOTE claims denial prediction model with SHAP explainability, achieving 89% precision on high-risk denial flags and informing appeals strategy. Designed FHIR-compliant data preprocessing pipelines using Python (Pandas, NumPy) and SQL for structured and semi-structured healthcare datasets (EHR, claims, patient records). Deployed models on AWS SageMaker with MLflow experiment tracking and model versioning, reducing model retraining cycle time from 2 weeks to 3 days via pipeline automation. Developed risk stratification models (Random Forest, XGBoost, Logistic Regression) to identify high-risk patient populations, improving preventive care planning accuracy by 22%.

Education

M.S. Computational Data Science at University of California, Riverside
September 1, 2024 - December 1, 2025

Qualifications

Add your qualifications or awards here.