By 2030, 80 million Americans will be over 65. We're building the infrastructure to keep grandma home.
Zingage builds AI agents that automate back-office operations for the largest home care companies in America — intake, scheduling, care coordination, compliance, patient engagement. All of it.
Adding 7 figures in ARR every month. Partnerships with the largest platforms in the industry. ~20 people. Based in New York City.
FIELD REPORT
The largest home care provider in the Pacific Northwest called us.
50+ branches across three states. 4,000 caregivers. 5,000 patients. Grandmothers recovering from hip surgery. Veterans with chronic conditions. Disabled adults who need daily help just to stay in their own homes. The people nobody builds software for.
Every day, 10,000 Americans turn 65. The demand for home care is growing faster than any other segment of healthcare — and the back office runs on phone calls, spreadsheets, and one coordinator who hasn't taken a day off in two years. Not because she's bad at her job. Because you can't hire and train coordinators as fast as the aging population grows. Human labor doesn't scale that way.
Monday, 2:14 PM
A hospital discharge planner calls about Margaret, 82, recovering from a hip replacement. She needs home care starting Wednesday — a nurse for wound checks, an aide for daily living. The discharge planner is calling five agencies. Whoever picks up first wins.
Before Zingage, this call goes to voicemail. The coordinator is on the other line. She'll call back in an hour, maybe two. By then, Margaret is someone else's patient.
Now the AI answers on the first ring. It captures the referral — diagnosis, insurance, physician, address, care needs. Checks caregiver availability against Margaret's location and schedule. Confirms placement. Four minutes. The discharge planner hangs up. Referral secured.
She hadn't taken a vacation in two years.
The scheduling coordinator manages 400 caregivers across the region. No backup. When she's out sick, the operation stalls. She isn't struggling because she's bad at her job — she's doing the work of nine people. She's their best person and there's only one of her.
Margaret needs five visits this week across two disciplines. The AI builds the schedule overnight. Matches a nurse by language and certification. Matches an aide by proximity and availability.
Thursday, 3 AM. An aide calls out sick. The agent picks up, checks 73 potential replacements against weekend availability, med training, Hoyer lift certification, gender preference, and a “do not call until 11 AM” restriction on one caregiver's profile. It assigns the replacement correctly to both ADL and IADL sub-shifts. Confirms with the caregiver. Updates the schedule. Notifies the family. Nobody was woken up.
“I actually ate lunch at my desk today instead of on the phone. That hasn't happened in months.”
— Scheduling Coordinator
3 weeks to close. 10 branches per week.
The deal closed in three weeks. An FDE flew to the first wave of branches. Four went live in week one: 1,979 calls answered, 2,632 scheduling issues resolved autonomously. That's 370 coordinator hours — the equivalent of nine full-time schedulers. Except you can't hire nine coordinators in a week. You can't even hire one.
The AI was live in seven days. Seven product improvements shipped in six — each one from observing the system in the field. A deaf client's profile wasn't flagged. An Oregon Medicaid authorization had an edge case. A caregiver had a “do not call until 11 AM” restriction that the old system didn't enforce. Fixed. Shipped. Live.
By week five, all 50+ branches were running. Missed clock-ins down 80%. Zero authorization expiration surprises. Audit time from hours to minutes. The AI catches things on Monday that humans used to catch on Friday.
“We've been trying to hire our way out of this for five years. You gave us nine coordinators in a week.”
— Regional Director
THE TEAM
The mad men and women of home care.
Ex-Uber, ex-Ramp, ex-Citadel. People who could be at any company in tech — and chose the hardest problem nobody else wants to touch. Not because they ended up in healthcare by default. Because they looked at 80 million aging Americans and decided someone has to build the infrastructure. Might as well be them.
TEAM FROM
BACKED BY
~20 people. Everyone either builds or sells — usually both. The GTM person writes SDR agents. The engineer presents to the customer's board. Nobody watches from the sidelines.
HOW WE WORK
CUSTOMERS FIRST
“Every line of code now has a patient on the other end. A race condition isn't an edge case — it's a missed visit.”
We deploy to customer sites. We shadow schedulers. We watch the system run. Then we ship improvements from what we observe.
VELOCITY
“7 product improvements in 6 days. Every one from sitting in a branch office watching the system run.”
Multiple enterprise customers per month. Two platform partnerships with the largest home care platforms in the industry launching.
IN-PERSON, IN THE ARENA
“You can't build for the real world from a laptop in your apartment.”
We work together in New York City. We fly to customer sites. The best ideas come from being in the same room — and from being in the field.
EXTREME OWNERSHIP
“We don't make excuses. We don't blame anyone or anything.”
You'll own entire deployments end-to-end. Fly to the customer, build the integration, present to their board, ship product from the field.
OPEN POSITIONS
Founding Forward Deployed Engineer
Deploy AI agents to enterprise customers. Build integrations on-site. Ship product improvements from the field.
$140-200K + 0.3-0.5% equity · New York City
START THE CHALLENGE →Founding Talent Engineer
We treat talent the way we treat product — as an engineering problem. Build the systems, pipelines, and intelligence that find and evaluate world-class people at scale.
Details coming soon
Founding AI Engineer
Simulation, reinforcement learning, and memory. Build the intelligence layer that makes our agents learn from every deployment and get smarter over time.
Details coming soon