Skip to main content
LPLav Patel

Manager, Data and AI Architecture

Lav Patel

Architecting AI products for healthcare at KUMC. Shipped CareScriptMD — an AI clinical documentation platform built on Whisper, Claude, and Llama. 24 publications. EB-1 Extraordinary Ability.

EB-1 Extraordinary Ability
CareScriptMD
24 Publications
1,031 Citations
h-index 17
npj Digital Medicine Board
4CE Consortium
Databricks Certified

CareScriptMD

AI Product Shipped

24

Publications

EB-1

Extraordinary Ability

Lav Patel, Manager of Data and AI Architecture at KUMC
Lav Patel — Manager, Data and AI Architecture, University of Kansas Medical Center

About Lav Patel

Lav Patel is Manager, Data and AI Architecture at the University of Kansas Medical Center (KUMC). Most recently, he architected and shipped CareScriptMD — an AI-powered clinical documentation platform that converts real-time patient visit audio into structured medical notes using Whisper, Claude, and Llama, all within HIPAA-compliant infrastructure. He served as sole technical builder, navigating IRB approval and dual-track institutional compliance to clear the regulatory path for clinical AI deployment.

Since joining KUMC in 2016, Lav has risen through six roles — from distributed systems engineering to leading AI product development in regulated environments. He previously owned security and governance for 21M+ Medicare and Medicaid patient records with zero breaches, shipped federated query infrastructure across 50+ hospitals, and delivered a 47x performance improvement on a CDC/NIH/FDA-funded national data pipeline.

A recognized authority in biomedical informatics, Lav holds an EB-1 Extraordinary Ability designation from USCIS. He has authored 24 peer-reviewed publications with 1,031 citations and an h-index of 17, and serves on the editorial board of npj Digital Medicine. As a member of the 4CE Consortium — a Harvard-led federated network of 300+ researchers across 96 hospitals in 5 countries — he contributes to international COVID-19 EHR research.

Key Achievements

  • Built CareScriptMD — AI clinical documentation platform using Whisper, Claude, and Llama
  • COVID-19 data pipeline connecting 10 hospitals in 7 days (typical: 1.5 months)
  • PCORnet CDM: 233 hours → 5 hours (47x faster). CDC/NIH/FDA/PCORI funded
  • Secured 21M+ Medicare/Medicaid patient records with zero breaches
  • Shipped federated query infrastructure across 50+ hospitals — delivered in 15 days
  • EB-1 Extraordinary Ability designation — USCIS recognition for extraordinary scientific contributions

Recognition & Affiliations

  • EB-1 Extraordinary Ability — U.S. Citizenship and Immigration Services
  • Editorial Board Member — npj Digital Medicine (Nature Portfolio)
  • 4CE Consortium Member — Harvard Medical School, 96 hospitals, 5 countries
  • Published in: npj Digital Medicine, The Lancet Digital Health, JAMA Network Open, EClinicalMedicine, JAMIA

Work Experience

Manager, Data and AI Architecture

University of Kansas Medical Center

Kansas City, Kansas

  • Architected and shipped CareScriptMD end-to-end — an AI-powered clinical documentation platform that converts real-time patient visit audio into structured medical notes, eliminating hours of manual charting per clinician per day. Collaboration with Jeff Burns, MD (PI).
  • Designed a production-grade clinical AI system: real-time audio capture, OpenAI Whisper deployed as a custom MLflow serving endpoint on Databricks, and multi-model LLM orchestration (Claude, Llama) for structured note generation — all within HIPAA-compliant infrastructure.
  • Engineered a production ML pipeline on Databricks with auto-scaling GPU serving endpoints and hybrid model routing strategy (Claude for accuracy-critical notes, Llama for high-volume intake), reducing inference costs.
  • Navigated dual-track institutional compliance — secured IRB approval, drove security reviews across both university and hospital systems, and cleared the regulatory path for clinical AI deployment.

Senior Team Lead — Biomedical Informatics Software Engineer

University of Kansas Medical Center

Kansas City, Kansas

  • Spearheaded the AllofUs Heartland consortium as Lead Software Engineer — built OMOP CDM from scratch and made the codebase available across University of Missouri, University of Iowa, and UNMC.
  • Directed a 10-engineer cross-functional team building enterprise-scale biomedical informatics systems.
  • Architected a COVID-19 data pipeline connecting 10 hospitals in 7 days — a process that typically takes 1.5 months.
  • Drove a 20-publication research program as head of the Knowledge Discovery Team, contributing to the international 4CE Consortium.

Team Lead — Biomedical Informatics Software Engineer | Solutions Architect

University of Kansas Medical Center

Kansas City, Kansas

  • Delivered a 47x performance improvement on the PCORnet Common Data Model pipeline (233 hours → 5 hours) — a CDC, NIH, FDA, and PCORI-funded national data infrastructure.

Biomedical Informatics Software Engineer | Data Engineer

University of Kansas Medical Center

Kansas City, Missouri

  • Re-architected core data pipeline systems to significantly improve speed and flexibility.
  • Optimized the department's main ETL process, decreasing run time from 28 days to 7 days.
  • Debugged and optimized i2b2, an open-source clinical data warehouse, achieving sub-second query performance.

Senior Enterprise Systems Engineer | DevOps Engineer

University of Kansas Medical Center

Kansas City, Missouri

  • Owned security and governance for 21M+ Medicare and Medicaid patient records, enforcing NIST-850, HIPAA, and CIS benchmark standards — zero breaches.
  • Shipped federated query infrastructure across 50+ hospitals in 15 days, delivering instantaneous cross-institutional results.
  • Drove Infrastructure as Code adoption (Terraform, Azure, Docker, K8s, Ansible, Jenkins).

Enterprise Systems Engineer | Distributed Backend Systems

University of Kansas Medical Center

Kansas City, Missouri

  • Implemented distributed systems architecture (Spark on Azure) to reduce hospital data loading from 2 days to 4 hours.
  • Deployed REDCap with custom add-on software, streamlining hospital resident evaluations from a week to instantaneous results.
  • Spearheaded golden OS image creation, reducing server build time from 1 month to 10 days.

Research Collaborations

Lav Patel is a member of the 4CE Consortium, an international federated research network that enables COVID-19 EHR analysis across institutions without sharing patient data.

4CE Consortium

Harvard Medical School — Lead Institution

International federated COVID-19 research network. 300+ researchers. 96 hospitals. 5 countries. Federated EHR analysis without patient data leaving institutions.

Americas Network

USA & Canada

PCORnet and i2b2 sites across academic medical centers. KUMC contributes data infrastructure leadership and pipeline engineering to the U.S. cohort.

European & Asia-Pacific

UK, France, Germany, Singapore & more

Cross-continental federated analysis sites contributing real-world EHR data to harmonized COVID-19 research studies.

Publications

24 publications · 1,031 citations · h-index 17 · i10-index 21

View full list on ORCID

npj Digital Medicine · 2021

Federated EHR Analysis of COVID-19 Clinical Outcomes Across International Hospital Networks

4CE Consortium
Federated Learning
COVID-19

The Lancet Digital Health · 2021

International Real-World Evidence for COVID-19 Research via Federated Electronic Health Records

Real-World Evidence
EHR
International

JAMA Network Open · 2022

Clinical Characterization of COVID-19 Patient Populations Using Multi-Site EHR Data

Clinical Informatics
COVID-19
Multi-site

JAMIA · 2022

Optimization of PCORnet Common Data Model for Distributed Clinical Research at Scale

PCORnet CDM
Data Infrastructure
Performance

Scientific Reports · 2023

Machine Learning for Clinical Outcome Prediction Using Federated Hospital Data

Machine Learning
Federated Learning
Outcomes

EClinicalMedicine · 2022

Temporal Trends in COVID-19 Outcomes Across Multi-National EHR Networks

COVID-19
Temporal Analysis
4CE Consortium

Representative selection. Full publication list available on ORCID and Google Scholar.

Technical Skills

Programming Languages

Python
R
SQL
Java
C++
Bash

AI & Machine Learning

PyTorch
Scikit-learn
CUDA
Triton
Clinical NLP
Deep Learning

Data Infrastructure

Apache Spark
Airflow
Databricks
Hadoop
Kafka
dbt

Cloud Platforms

Microsoft Azure
AWS
Docker
Kubernetes
CI/CD

Clinical Standards

OMOP CDM
PCORnet CDM
i2b2
FHIR
HL7

Databases

PostgreSQL
MySQL
Oracle
SQL Server
MongoDB

DevOps & Tools

Git
Jenkins
Linux
Bash
REST APIs

Education & Credentials

Louisiana Tech University

Bachelor's and Master's Degrees

BS and MS from Louisiana Tech University.

Certifications

Databricks & Azure

  • Azure Databricks Platform Architect
  • Databricks Certified Data Engineer Associate

EB-1 Extraordinary Ability

U.S. Citizenship and Immigration Services

USCIS EB-1 Extraordinary Ability designation — reserved for individuals who have risen to the very top of their field. Awarded in recognition of Lav Patel's sustained national and international acclaim in biomedical informatics and healthcare AI.

Frequently Asked Questions

Who is Lav Patel?

Lav Patel is Manager, Data and AI Architecture at the University of Kansas Medical Center (KUMC), where he architected and shipped CareScriptMD — an AI-powered clinical documentation platform. He is an EB-1 Extraordinary Ability holder with 24 peer-reviewed publications, over 1,031 citations, and an h-index of 17. He serves on the editorial board of npj Digital Medicine and is a member of the 4CE Consortium, a Harvard-led federated research network spanning 96 hospitals across 5 countries.

What is Lav Patel's area of expertise?

Lav Patel specializes in building AI products for healthcare, clinical data architecture, and federated data networks. He architected CareScriptMD, a production AI clinical documentation platform using Whisper, Claude, and Llama. His expertise spans OMOP and PCORnet Common Data Models, clinical NLP, real-world evidence generation, and large-scale EHR data pipelines. Over six roles at KUMC, he has progressed from distributed systems engineering to leading AI product development in regulated environments.

What has Lav Patel published?

Lav Patel has authored 24 peer-reviewed publications with 1,031 citations, h-index 17, and i10-index 21. He has published in npj Digital Medicine, The Lancet Digital Health, JAMA Network Open, EClinicalMedicine, Scientific Reports, BMJ Open, and JAMIA. His research focuses on COVID-19 data science, federated learning for healthcare, and clinical informatics.

What is the 4CE Consortium?

The 4CE (Consortium for Clinical Characterization of COVID-19 by EHR) is an international federated research consortium led by Harvard Medical School. It connects 300+ researchers across 96 hospitals in 5 countries to analyze anonymized EHR data for COVID-19 research without sharing patient data across institutions. Lav Patel is a member and data infrastructure contributor.

How can I contact Lav Patel?

Lav Patel can be reached by email at lavpatel39@gmail.com. He is active on LinkedIn at linkedin.com/in/lav-patel/ and on X (Twitter) at @lav_p_patel. His research profile is available on ORCID at orcid.org/0000-0002-8626-137X.

Get in Touch

Interested in collaborating on healthcare data research, AI infrastructure, or speaking engagements? Reach out directly.