18+ years solving the hardest problems in bioinformatics — from raw sequencing reads to clinical-grade insights. I built OmniBioAI because fragmented, hard-to-reproduce tools were slowing down real science.
Built solo, out of necessity
I built OmniBioAI alone. No team, no funding rounds — just 18 years of accumulated frustration with the state of bioinformatics tooling and a DGX Spark to make it real.
After working across genomics, proteomics, and drug discovery pipelines at hospitals, universities, and genomics labs spanning India, Malaysia, Qatar, Saudi Arabia, and the United States, I kept hitting the same walls: tools that broke between clusters, pipelines that couldn't be reproduced, AI that couldn't reason about biology.
"Every lab was rebuilding the same infrastructure in isolation. I decided to build the platform I always wished existed — and ship it as a single desktop install."
What started as a personal toolkit is now a production system: 2.1M lines of code, 28 microservices, 36M PubMed abstracts indexed, 600+ workflow bundles, and 1 PetaFLOP of AI compute — built by one person over three years.
The mission
Modern bioinformatics is fragmented. A graduate student's analysis breaks when they move to a new HPC. A clinical team revalidates pipelines every time a tool version changes. Literature search is disconnected from computation. AI assistants hallucinate biology.
OmniBioAI is my answer: a reproducible, AI-native operating system for biomedical research that runs identically on a laptop, a Slurm cluster, or AWS — with a 9-stage RAG pipeline over 36M PubMed abstracts, full LIMS integration, and provenance from raw FASTQ to final biological insight.
Currently a Bioinformatics Programmer and Research Associate at the University of Kansas Medical Center (KUMC), I build OmniBioAI in the margins — nights, weekends, and every spare hour. It is a solo project in the truest sense.
Career timeline
Skills & stack
Peer-reviewed research
Interested in collaborating?
Whether you're a researcher, institution, or biotech — I'd love to hear about your multi-omics challenges.