Healthcare · AI · 2024 – now · Founder
Parahealth
Prior authorization eats roughly 13 hours a week per clinician. Parahealth automates the gather → submit → follow-up loop end to end — AI reads the chart, fills the payer form, and chases the status.
Problem
Prior authorization is one of the biggest sources of administrative burden in US healthcare: clinics chase faxes, sit on hold with payers, and re-key the same clinical data into a dozen different portals.
It delays care, burns out staff, and is almost entirely undifferentiated work — exactly the kind of thing software should eat.
What I built
An end-to-end platform: it ingests the relevant chart context, determines payer-specific requirements, drafts and submits the authorization, and then monitors and follows up on status automatically.
Built the product from zero — data model, the AI pipeline that reads charts and fills forms, the reviewer UI, and the integrations.
Stack
Outcomes
- Collapses a multi-day, multi-touch workflow into minutes of human review.
- In active use; founder-led.