Looking for a job

Prashanth Dhanasri

@prashanthdhanasri

Message

I build scalable LLM, RAG, and voice AI systems for production workflows.

United States

Message

What I'm looking for

I'm looking to build production AI systems where LLM applications, RAG, agentic workflows, evaluation, and reliable deployment solve meaningful product problems in regulated or high-scale environments.

At Mastercard, I build stateful LLM applications for claims and payment-support workflows, handling over 10,000 daily interactions. I use context engineering, RAG, Azure OpenAI, and agentic workflows to improve retrieval coverage and resolution accuracy.

I've built fraud-detection models, distributed Spark pipelines, and containerized ML services at HHAeXchange, processing more than 10 million transactional records daily. My work improved anomaly identification accuracy, reduced false positives, and shortened deployment time through CI/CD automation.

At Amazon, I developed RAG-based product-discovery systems and real-time voice capabilities with SIP and SBC optimizations. I focus on rigorous evaluation, latency, privacy controls, observability, and reliable deployment for production AI systems.

Experience

Work history, roles, and key accomplishments

Current

GenAI / ML Engineer

Current

Mastercard

Oct 2025 - Present (11 months)

Built stateful LLM applications with context engineering and RAG, improving retrieval coverage by 20%. Designed agentic workflows and CI/CD pipelines, reducing deployment errors by 35% and enhancing security and observability.

LLM Context Engineering RAG Agentic Workflows Azure OpenAI Telephony CI CD GitHub Actions Python Docker Kubernetes fastAPI Lang Chain LangGraph Observability Security Privacy DSPy GEPA

Machine Learning Engineer

HHAeXchange

May 2024 - Sep 2025 (1 year 4 months)

Developed deep learning and machine learning fraud-detection models, improving anomaly identification accuracy by 18%. Built scalable data pipelines and deployed containerized ML services, reducing deployment time by 40%.

Deep Learning Machine Learning Fraud Detection Spark Docker Kubernetes CI CD Lang Chain Hugging Face Transformers Statistical Measurement Error Analysis

AI/ML Engineer

Amazon

Jan 2022 - Dec 2023 (1 year 11 months)

Designed distributed data stores and scalable pipelines, improving data retrieval latency by 25%. Developed RAG-based LLM applications and voice path optimizations, reducing service response time by 30%.