Developed a >$1M ARR product for quick data delivery to clients, architecting a scalable backend solution that enables rapid healthcare data access and processing.
Healthcare AI platform built for trust and reproducibility — not just the right answer, but the right answer with proof. Marmot classifies each question and routes it to the right model per subtask, versions every intermediate step so an analysis reproduces quarter-to-quarter, and gates every output through healthcare-specific evaluations before delivery. Ships as Navigator (plain-language analytics — multi-month analyses in minutes, with the reasoning behind every answer) and the Marmot Development Kit (the Healthcare Map directly in Snowflake, Databricks, or Python — no data exports). Built on top of Komodo’s Healthcare Map.
Beauty-clinic SaaS for Indonesian clinics — appointments, customer records, services, sales & marketing, reporting, and a public booking menu, all in a fully localized Indonesian-language UI. React 18 frontend, FastAPI backend, Clerk auth.
AI-powered image generation platform that edits marketing images using Google Gemini AI. Features include uploading photos for editing, adding prompts to modify images, powered by the new gemini flash nano banana model.
Web application for splitting itemized restaurant bills fairly among friends and colleagues. Features AI-powered receipt parsing with Claude AI, receipt-style interface, smart proportional bill splitting, and comprehensive analytics tracking.
AI-powered restaurant CRM that replaces manual data entry, and segments the customer base to increase revenue
Created a Streamlit-based platform that enables 10x faster prototyping for non-frontend developers, dramatically improving development velocity and reducing time-to-market.
Standardized TypeScript configurations across 20+ packages in Turborepo-managed monorepo and established Cypress E2E testing framework, significantly reducing production bugs.
Designed and developed big-data visualization dashboards using D3.js for genomics applications, enhancing clinical data insights and enabling researchers to better understand complex genomic datasets.
Published machine learning research developing an automated CRISPR gRNA design platform for crop genomes. Created improved scoring algorithms using Support Vector Regression that increased prediction accuracy by 19% over existing tools, particularly for A/T-rich and polyploid crop genomes like soybean and Miscanthus.
I'm always open to discussing new opportunities and exciting projects. Feel free to reach out at [email protected].