● 3 projects

Selected work

Selected work — AI systems, healthcare tooling, and data platforms I have built and shipped.

Claraly: Physician co-pilot

01

Claraly: Physician co-pilot

Physician co-pilot that saves clinicians time so they can focus on practicing medicine.

Challenge: Physicians spend excessive time on administrative tasks like taking notes and coding, taking away from patient care.
Solution: Developed an AI co-pilot that generates clinical notes ambiently, handles dictation, recommends billing codes, and answers medical questions.
Outcome: Significantly reduced physicians' administrative workload, allowing them to focus more on patient care.

AI
Healthcare
LLM
React
NodeJS

Healthcare Knowledge Graph

02

Healthcare Knowledge Graph

The Healthcare Knowledge Graph RAG with Neo4j, LangChain, and Llama 3 for enhanced clinical decision support.

Challenge: Traditional RAG systems struggle with broad questions about entire medical text corpora that require query-focused summarization.
Solution: Developed a Knowledge Graph RAG system using Neo4j, LangChain, and Llama 3 to improve comprehension of medical documents.
Outcome: Significantly improved the comprehensiveness and diversity of answers compared to basic RAG systems.

Neo4j
LangChain
Llama 3
Healthcare
AI

Clinical Data Platform

03

Clinical Data Platform

Collecting data from various EHR vendors and building a data warehouse following the Medallion Architecture.

Challenge: Data ingestion bottlenecks were delaying client onboarding due to inefficient processing of diverse healthcare data formats.
Solution: Created an end-to-end data pipeline using the medallion architecture with Azure Data Pipeline, Databricks, and Synapse.
Outcome: Reduced client onboarding time from weeks to days by automating data ingestion processes.

Azure
Databricks
SQL Server
Healthcare
Data Engineering