New startup ideas · AI and software · Vertical AI agents
startup idea
Clinos
The certified clinical core that every consumer health agent builds on.
Clinos is a consumer subscription: a personal health agent that holds your records and wearable data and manages chronic-care routines, medication adherence, and post-discharge recovery.
- Software subscription
- Consumer
- $10-100B market
- Platform others build on
- Global from day one
3/5
venture judge
51
similar startups, last 2 years (242 all-time)
97%
of 3 nearest real companies still alive
yes
8 matching federal grants and programs
Direction supported by government programs and grants
Test it before you build it
$900 · 6 weeks · 15 prospects
For $900 and 6 weeks, learn whether US hospitals paying CMS readmission penalties will sign dated $20,000 pilot commitments before any certification exists.
Riskiest assumption · A care management director at an HRRP-penalized US hospital will commit to a dated, signed $20,000 pilot with an uncertified startup because each avoided readmission is worth $10,000 or more in payment and penalty relief, since without that channel demand the certification-first strategy repeats Galen AI's death with extra years added.
1Focus group: who and where
The director of care management or population health at a US hospital carrying a CMS Hospital Readmissions Reduction Program penalty on heart failure and AMI readmissions, whose FY2026 penalty just posted and who owns the readmission number this quarter.
where to find 15 · The public CMS HRRP FY2026 penalty file on CMS.gov, filtered to hospitals near the 3 percent maximum penalty, a ready-made list with hospital names; local chapter meetings of the American Case Management Association (ACMA), where exactly these directors gather; warm introductions through cardiologists and discharge-planning nurses, the one channel hospital buyers reliably answer.
2Sell first, build later
A 90-day nurse-supervised post-discharge follow-up pilot for heart failure and AMI patients on one cardiac unit: daily automated check-ins with RN escalation under the hospital's existing protocols, weekly readmission reporting, starting Q1 2027.
the ask · $20,000 per pilot, invoiced at contract signature
a real yes · A real yes is a signed LOI naming the unit, the start month and the $20,000 fee, followed by a contract in legal review; a verbal 'the CMO loves it', an unpaid pilot, or an innovation-team meeting does not count.
3Small experiments
The first one attacks the riskiest assumption; each ends with a number that says whether to run the next.
1. Penalty-hospital pitch calls
$600 · 21 days
Pull the CMS HRRP penalty file, shortlist 40 hospitals with the highest penalties within driving distance, and book 15 calls with care management directors through ACMA chapter meetings and clinician introductions. Pitch a $20,000, 90-day, nurse-supervised follow-up pilot on one cardiac unit and ask for a written scope review, not a demo.
keep going if · 5 of 15 directors request the written pilot scope
2. LOI with dated start
$200 · 21 days
Send each interested director a two-page pilot scope reviewed by an RN advisor, plus a letter of intent naming the $20,000 fee, the unit, and a Q1 2027 start month, signed electronically. The founder walks each one through it on a 20-minute follow-up call and asks for signature within two weeks.
keep going if · 3 signed LOIs with a named unit and start month
3. Module developer probe
$100 · 10 days
Call 10 founders of condition-specific consumer health apps, found in the App Store health category and the Y Combinator directory, and ask what regulatory clearance costs them today and whether they would ship on a certified core for a 20 percent platform take. This prices the platform half of the thesis before any certification spend.
keep going if · 4 of 10 say clearance is their blocker and ask for API documentation
4Collect a deposit up front
Tesla took $1,000 refundable reservations for the Model 3 and $100 for the Cybertruck before building either: the deposit is the measurement, not the revenue.
$0
no cash yet: take a signed commitment
how · A signed LOI with a dated start, because a US hospital cannot credibly move money to an uncertified vendor before compliance, legal and a BAA; the care management director signs a letter naming the $20,000 fee, the cardiac unit, and the Q1 2027 start month, and that signature is the measurement.
what it reserves · One of three Q1 2027 design-partner slots, the named cardiology unit, and the $20,000 pilot price locked through Q2 2027.
refund · No money moves at signature; either side can withdraw in writing any time before the pilot contract is signed.
target · 3 signed LOIs with dated starts from 15 conversations within 45 days
before taking money · Do not give patients clinical advice, touch PHI without a signed BAA, or market the service as a diagnostic before counsel settles the FDA device pathway; the pilot must run as nurse-supervised follow-up under the hospital's existing protocols.
Go: build it if
3 of 15 hospitals sign a dated LOI and at least 1 pilot contract enters legal review within 6 weeks, plus 4 of 10 module developers name clearance as their blocker.
Kill: stop if
0 signed LOIs and fewer than 3 scope-review requests after 15 pitches; if penalty-paying hospitals will not even commit on paper, the insurer and pharmacy channels further from the pain will not either, and the multi-year certification road has no buyer at the end.
5 Scripts to run itoutreach message, landing copy, deposit terms · click to open
outreach message
Your hospital is on the FY2026 CMS readmission penalty list, and every heart failure patient who bounces back inside 30 days costs you roughly $15,000 in lost payment and penalty. I'm building Clinos, a nurse-supervised AI service that follows every cardiac discharge daily for 30 days: meds, symptoms, escalation straight to your team. We're signing three design partners for a $20,000, 90-day pilot on one unit starting Q1 2027. Do you have 20 minutes this week to review the protocol?
landing page
Cut cardiac readmissions with AI follow-up your nurses supervise $20,000 pilot: 90 days, one cardiac unit, every discharge followed daily for 30 days with weekly readmission reporting. Book a scoping call to reserve one of three Q1 2027 design-partner slots.
deposit terms
This letter reserves one of three Q1 2027 design-partner slots for a $20,000, 90-day pilot on the cardiology unit you name, with a start date we set together in writing. No money moves until your compliance and legal teams sign the pilot contract, and either side can withdraw in writing before then. The $20,000 pilot price is locked for signatories through Q2 2027.
Would you run this test?
One tap. The yes-share feeds the Demand pillar of this idea's score; nobody sees who answered.
Budgets are out-of-pocket estimates for a team of one to three, US market. Size the deposit to the deal, and check the terms before taking money in a regulated line.
Scorecard
One score that balances how trendy the idea is, the demand for it and its potential for 100x, with competition measured relative to every other idea in the catalog. Recent startup trends first, government priorities second.
50
Idea Score, 0-100 · raw 31.1 x 1.61
Crowded
competition: more crowded than 98% of ideas · headwind x0.51
+4.8
government priorities, secondary (163 matching grants)
Trend
54
Is the wave forming now? 2025-26 entrants vs 2023-24, rounds since 2025, the sector's live-batch direction, the 2026 trend analyst.
- Entrants 2025-26 vs 2023-24 (similar companies)67
- Rounds announced 2025+ in the sector25
- Sector direction (live batch)100
- 2026 trend analyst25
Demand
42
Does anyone want it? YC's current RFS, companies already paid for something similar, the operator judge, founders' yes-rate in decks, readers who would run the test.
- YC asks for it (current RFS: idea / sector)0
- Someone already pays (similar companies, recent / all-time)100
- Operator judge: real pain25
100x potential
60
Can it return a fund? The venture judge (double weight), market-size and moat axes, neighbours still alive, the technologist judge.
- Venture judge50
- Market size axis67
- Moat axis80
- Neighbours still alive63
- Technologist judge50
Score = 100 x cbrt(Trend x Demand x 100x) x (1 - 0.5 x crowding) + government bonus (max 5), calibrated so the 95th-percentile idea scores 90 (order never changes). A geometric mean: a weak pillar cannot be papered over. Percentiles are among the 272 ideas in the catalog; the terms matched were certified, clinical, core, consumer, health, builds, subscription, personal.
The idea in full
- What
- Clinos is a consumer subscription: a personal health agent that holds your records and wearable data and manages chronic-care routines, medication adherence, and post-discharge recovery. Underneath is a clinically validated reasoning core carrying the medical-device certification, on which third-party developers ship specialty modules for cardiac rehab, diabetes, or physio without seeking their own clearance. Distribution runs through insurers and pharmacy chains who bundle the subscription because it demonstrably cuts readmissions.
- Why now
- The NSF is funding an I-Corps award (~$50k) on the translation potential of conversational AI for post-discharge patient follow-up, Prescience, Inc. (yc S26) pitches healthcare for the age of abundant intelligence, and graveyard entry Galen AI (2025) proved a standalone consumer health agent app dies without clinical credibility or a distribution channel.
- Wedge: first customer and entry point
- Post-discharge follow-up for one cardiac condition, validated in a single-hospital study, sold through one European insurer who pays because each avoided readmission is worth thousands.
- Path to 100x
- Consumer digital health is a tens-of-billions market, and the platform mechanism is the multiplier: once the certified core exists, hundreds of specialty developers extend it into conditions Clinos would never build itself, each paying a platform take. The certification plus insurer channel contracts form a gate a newcomer needs years of clinical validation to pass, so the category consolidates onto one core.
- Ceiling
- If insurers negotiate the bundle price down to a wellness-app level, per-user economics never support a $1B outcome despite the reach.
- Closest real companies, as the generator saw them
- Galen AI (dead) was a standalone app with no certification or channel; Prescience, Inc. is provider-side; Cova does caregiving finance, not clinical reasoning. Clinos is the certified substrate others build on, not another app.
- Main risk
- Regulators decide every third-party module is a separate medical device, destroying the platform leverage that justifies the slow certification road.
Five judges
Each judge scores every idea in the catalog with a named rubric; the venture judge decides whether a card is shown at all (4-5 is venture-grade).
Venture investor
3/5
Certified core others build on is real platform leverage, but insurer-negotiated consumer pricing is exactly the per-user cap the card admits.
Bootstrapper
1/5
Medical-device certification plus insurer contracting is a multi-year regulatory path with no revenue underneath it.
Operator
2/5
Insurers pay for avoided readmissions, but device certification plus third-party module developers puts real revenue three years and one regulator decision away.
Technologist
3/5
The clinically validated core is genuine, but platform leverage hinges on regulators treating third-party modules as exempt, a permission rather than an engineering asset.
Risk
3/5
Device certification is strong footing, but one regulator ruling that each third-party module needs its own clearance destroys the entire platform premise.
trends
2/5
A $50k NSF I-Corps award is the only dated hook; the certified-core play was equally available in 2023, with healthcare grants worth one point.
Similar startups in the directory
Companies whose pitch matches most of the idea's terms (certified, clinical, core, consumer, health, builds, subscription, personal): 242 all-time, 51 from the last two years. Same matching as Idea Check.
AI-powered scalable infrastructure + mental health services, starting…
Powering preventative care delivery for Medicare specialties
We are building an AI-native telemedicine clinic for addiction. 🌻
Smart packaging that tracks consumer product usage
AI-native CRO for clinical trial operations
AI-native patient engagement
Function Health for Pets.
AI-powered sleep care, covered by insurance.
Hydrostasis is a developer of hydration monitoring systems.
Personal AI healthcare agent, powered by clinical and wearable data.
Voice AI for medical language translation
The AI communication layer for healthcare
The generator's reference companies
Real companies the model named as closest when it wrote the card, with their fate. A check mark is a company the radar could verify in its directory.
Public money in this direction
US federal grants, SBIR/STTR awards and open opportunities from the radar's public-money feed, matched to the idea's terms; the sector totals give the context.
8
grants and programs matching the idea
72
startup-relevant grants in Vertical AI agents
$135M
awarded in the sector, tracked
- Cardiovascular Safety and Scalable Manufacturing for Investigational New Drug Filing of a Novel Oral Drug to Accelerate Musculoskeletal Recovery in Older Adultsawardhigh relevance
NIH / NIA · SBIR phase II · $2M · posted 2026-09-23
NIH / NIA · SBIR phase II · $889K · posted 2026-09-22
- Automated MRI-Based Musculoskeletal and Body Composition Assessment for Healthy Agingawardhigh relevance
NIH / NIA · SBIR phase II · $1M · posted 2026-09-18
- Daily Burn Active Together: A personalized, dyadic exercise and health program for people living with dementia and their family care partnersawardhigh relevance
NIH / NIA · SBIR phase I · $500K · posted 2026-09-17
NIH / NIDDK · SBIR phase II · $325K · posted 2026-09-15
NIH / NINDS · STTR phase I · $322K · posted 2026-09-10
- Stasys ATLAS platelet test: Expanding test cartridge manufacturing under ISO 13485 regulatory controls for FDA trial submissionawardhigh relevance
NIH / NHLBI · SBIR phase I · $326K · posted 2026-09-01
NIH / NHLBI · SBIR phase I · $320K · posted 2026-09-01
Market signal
What the radar sees in Vertical AI agents: new companies by cohort year, the forming YC batch, and outcomes since the February snapshot.
Vertical AI agents · 159 → 261 → 364 → 411 → 353 new companies 2022 → 2026 · 93% aliveYC F26 live: 25 in this cluster, 20% of the batch (was 25% in S26)Since February, of 698 YC companies here: 19 acquired, 9 shut down, 219 rewrote their pitch
Design attributes
The card is one cell of a designed set: every axis below was chosen before the text was written, and the text had to realize it.
- Buyer
- Consumer
- Business model
- Software subscription
- Path to 100x
- Platform others build on
- Market size
- $10-100B market
- Capital intensity
- Capital-medium (ops, field teams)
- Speed to revenue
- R&D first, revenue after 3 years
- Technical depth
- Deep tech: ML, hardware, bio
- Go-to-market
- Partners and channels
- Moat
- License or regulatory moat
- Geography
- Global from day one
- Regulation
- Some regulation
- Vibe
- Hot space
Listed under
An idea sits in its own sector and in any sector its text clearly touches.
More ideas like this
AI and software · Vertical AI agents
Bidcraft
Small contractors upload plans and get a priced, win-rate-tuned quote in twenty minutes.
Bidcraft is a self-serve subscription for specialty trade subcontractors and field service firms: drop in a PDF plan set, a site photo set or an inspection report, and the system does takeoff, prices materials from live supplier catalogs in the user's country, applies the firm's labour rates and returns a client-ready quote.
AI and software · Vertical AI agents
Fiscala
A licensed AI tax agent that files for cross-border workers in every country they touch
Fiscala is an AI agent that acts as the registered tax preparer for expats, remote workers, and digital nomads, ingesting payslips, broker statements, and residency records to prepare and file returns in multiple jurisdictions at once.
AI and software · Vertical AI agents
Prybar
An API that lets any AI agent operate the legacy Windows software small businesses actually run.
Prybar sells small businesses a small on-premise appliance plus an API: it drives the closed, keyboard-only applications that run auto body shops, repair companies, restaurants and brokerages, exposing read and write operations as typed endpoints with rollback and an audit trail.
AI and software · Vertical AI agents
Ironfold
A plug-in appliance that runs AI agents inside regulated enterprises at a tenth the cost.
Ironfold builds a rack appliance that hospitals, insurers and banks order online, plug into their network, and use to run vertical AI agents on premises, with audit logs and patient or financial data never leaving the building.
AI and software · Vertical AI agents
Coderoot
Self-serve AI agents that maintain legacy enterprise code for a tenth of outsourcing cost.
Coderoot's agent connects to an enterprise's legacy estate - ABAP, COBOL, aging Java, ERP customizations - builds a behavioral map of the system, then works the maintenance backlog: patches, upgrades, regression tests and documentation.
AI and software · Vertical AI agents
Bondera
The underwriting and bonding layer that makes AI agent work trustworthy.
Bondera is an API that agent vendors embed to attach a performance guarantee to their agents' output: if the agent misfiles a return, botches an invoice, or mishandles a claim, Bondera pays the customer.
Fictional company written 2026-08-26 from MarkosWeb data; the companies, grants and numbers around it are real and tracked. Treat the idea as a research prompt, not a plan.