Company profile · 4 min read
Kaigo Health is a human-and-AI business office for skilled-nursing operators
Kaigo Health helps skilled-nursing facilities find leakage, capture reimbursement and execute administrative recovery workflows.
Cofounder of Lazyweb (a16z Speedrun). Previously grew subscriptions and built native ads at Duolingo. Writes First1000 (90,000+ subscribers).
Published · Updated
Kaigo Health’s current product is less “AI outpatient workforce” than a super-powered business office for skilled-nursing facilities. That sharper description makes the buyer fit much easier to see.
What it does
Kaigo’s homepage says its human-plus-AI team finds money leaking from SNF operations and helps recover or capture it. The workflows include pharmacy and vendor charges, reimbursement and payer dollars, authorizations, appeals, referrals and disputes. The company says it starts with exports, then works across invoices, P&L, census, payer and vendor data. Kaigo Health homepage
The site gives one concrete case: at a 34-bed, five-star building, the company says a pharmacy invoice fell from $30,000 to $7,000 and that average recovery was $6,700 per month. It also lists other company-reported findings around discharge costs, unused subscriptions, missed skill-in-place revenue and food spend. Those are company case claims, not an independent audit.
The YC profile describes the earlier framing as an AI-powered outpatient care-management platform for Medicare follow-ups, check-ins and monitoring.
Why I’d look closer
The current positioning is practical. Skilled-nursing operators do not need another general healthcare chatbot. They need to find money in fragmented billing, vendor and payer workflows, then get the paperwork or appeal moving.
The founders’ personal reason to care is also unusually specific. The YC source describes the team seeing missed medications, readmissions, isolation and staff shortages in older-adult care, and records more than 600 hours working with older adults. It also describes backgrounds connected to Neuralink, Stanford, NVIDIA and healthcare AI.
That context helps explain why Kaigo is selling operational recovery, not just patient conversation.
What could make it the wrong choice
The workflow depends on access to sensitive operational data and on the facility’s staff reviewing actions. The homepage explicitly says Kaigo prepares analysis and drafts while prescribers, billers and staff review, decide and submit. That boundary is a feature, but it also means the buyer needs clear ownership for approvals.
Pricing is not published in the sources checked. A facility should ask how recovered dollars are defined, how the baseline is calculated, which payer and vendor workflows are supported, how corrections are handled, and what data remains after the engagement.
My editorial take
I would shortlist Kaigo for a skilled-nursing operator that can export the right financial and census data and has a business-office owner who will act on the findings. I would not buy the broader “AI workforce” idea in the abstract. The practical wedge is narrower: find one recurring leakage pattern, recover it with a human-approved workflow, and prove the dollars belong to the facility.
Quick facts
| Field | Sourced detail |
|---|---|
| Product | AI and human business-office support for SNFs |
| Buyer | Skilled-nursing and post-acute-care operators |
| Published proof | Company case claims around pharmacy, payer, vendor and reimbursement recovery |
| Pricing | Not published in the checked sources |
| Main fit question | Can the facility provide data and own the approval loop? |
Sources checked
Kaigo Health homepage and YC company profile were checked on 2026-09-19.
Cohort context
Kaigo Health is listed in Fall 2025. In our 2026-09-18 directory snapshot, 11 of 146 listed companies in that cohort have YC’s primary industry label Healthcare (7.5%). This is a current-directory comparison, not an original intake count or a performance ranking. Nine-cohort dataset.
Public website snapshot
Observed 2026-09-19T16:15:03.962Z in raw homepage HTML. This records visible metadata and advertised links, not agent execution or product quality.
| Signal | Homepage observation |
|---|---|
| Product description metadata | Observed |
| Canonical link | Observed |
| H1 or H2 heading | Observed |
| Typed structured data | Not observed in this response |
| Docs/developer link | Not observed in this response |
| Pricing link | Not observed in this response |
| llms.txt link | Not observed in this response |
| Markdown alternate | Not observed in this response |
Public observations · Collection method. Missing links here do not establish that a capability or file is absent elsewhere.
About the author
Cofounder of Lazyweb (a16z Speedrun). Previously grew subscriptions and built native ads at Duolingo. Writes First1000 (90,000+ subscribers).
I cofound Lazyweb and publish Mudpie. This is an owner-written publication, not an independent testing organization. Research notes distinguish observations, sourced reporting and editorial judgment.
First1000 ↗ · X ↗