Open to opportunities

Ash Kanwal

@ashkanwal

Message

Senior AI Engineer specializing in LLM RAG systems and scalable ML pipelines for enterprise impact.

United States

Message

What I'm looking for

I’m looking to build production LLM/RAG and scalable ML systems—focused on measurable retrieval quality, efficient pipelines, and reliable deployment—so teams can move faster from complex data to business-ready outcomes.

I’m a Senior AI & Machine Learning Engineer with 6+ years delivering production-scale intelligence systems across Healthcare, LegalTech, and Global Enterprise. I architect advanced LLM solutions and scalable ML pipelines that turn complex data into strategic business value and high-performance user experiences.

In my recent role, I built an enterprise-grade RAG system with LangChain and OpenAI, boosting retrieval accuracy by 14% and reducing time required for hands-on document review. I also designed high-throughput ML pipelines with BigQuery partitioning/clustering (35% runtime reduction, 20% lower query costs) and automated anomaly detection on AWS SageMaker—improving model accuracy by 25% and reducing churn by 12%.

Experience

Work history, roles, and key accomplishments

Current

Senior AI Engineer

Current

Smart Advocate

Jan 2024 - Present (2 years 5 months)

Identified a manual medical record review bottleneck accounting for 60% of the discovery-to-demand lifecycle, slowing client case evaluation and settlement timelines. Architected an enterprise-grade RAG system with LangChain and OpenAI, improving retrieval accuracy by 14% and reducing the time required for hands-on document review.

Lang Chain OpenAI Retrieval Augmented Generation (RAG)Semantic Search Intelligent Q&A High Dimensional Vector Similarity

Machine Learning Engineer

Midwestern

Jan 2022 - Jan 2024 (2 years)

Engineered high-throughput ML pipelines over millions of records using BigQuery partitioning and clustering, reducing runtimes by 35% and query costs by 20%. Built AWS SageMaker anomaly detection with Docker for 10K+ clinical notes monthly, improving model accuracy by 25% and reducing user churn by 12% via predictive insights, and implemented Airflow-based ETL orchestration with an 88% success rat

BigQuery AWS Sagemaker Docker Anomaly Detection Airflow ETL

AI Engineer

Epam Systems

Jan 2019 - Jan 2022 (3 years)

Supported a Tier-1 healthcare provider in resolving chronic integration delays caused by manual rewriting of data transformation scripts for each clinical model iteration. Built a modular Python framework to automate data transformation and parsing, reducing redundant code by 23% and accelerating time from data collection to results.

Python Data Automation Healthcare Data Integration AI ML Engineering