FutureClinic wants you to text your doctor - and get a signed answer today
An AI takes your medical history. A real physician reads it, writes your plan, and records themselves signing off. The waiting room is optional.
The typical way to see a doctor in America involves a phone tree, a three-week wait, a drive, a parking lot, a clipboard of questions you have answered a dozen times before, and roughly seven minutes of the physician's actual attention. FutureClinic, a company that came out of Y Combinator's Fall 2024 batch, looked at that sequence and asked which parts a patient would keep if they had the choice. Its answer is short: the doctor, and nothing else.
The New York startup runs a platform where any licensed physician can open a digital clinic and treat patients entirely through chat. There is no appointment to book. A patient picks a doctor, describes what is wrong in text, photos, or voice notes, and gets a reply - usually the same day. What sits between the message and the reply is where the company has placed its bet.
An AI does the intake. A human does the deciding.
Before a doctor ever opens a case, an AI the company calls the "Chief Resident" has already spoken with the patient. It conducts the medical history through an empathetic chat or voice interface, asks the follow-up questions a good clinician would, works through the likely diagnoses, and drafts treatment suggestions grounded in current medical guidelines. Only cases suited to virtual care get routed onward. The doctor receives a concise written summary rather than a blank chart.
Then the human takes over the part that matters. The doctor reviews the case, writes the plan, orders any prescriptions or labs, and records what FutureClinic has trademarked as a Human Signature - a short live video or audio note delivering the plan in their own voice. The AI never signs anything. In the founders' framing, the software handles roughly 99% of the work, and the physician does the high-value 1% that is theirs alone.
History-taking, diagnostic narrowing, guideline-based drafting, routing.
Reviews, decides, prescribes, and signs the plan on camera.
It is a pointed division of labor at a moment when much of the industry is racing to see how much of the physician an algorithm can replace. FutureClinic drew the line in the other direction: automate the clipboard, protect the judgment.
The distinction is not cosmetic. Intake is where most of a doctor's time leaks away - transcribing symptoms, chasing the timeline, ruling out the obvious, typing notes that no one enjoys typing. It is also, crucially, the part of medicine that is closest to pattern-matching, which is exactly what current AI does well. Diagnosis under uncertainty, weighing a patient's history against a hunch, and taking responsibility for a plan are not. By assigning each task to whichever agent is better suited, FutureClinic is trying to make the doctor faster without making them smaller.
Founders who both write the code and see the patients
FutureClinic was founded in 2024 by four people, and the unusual detail is that three of them are both physicians and software engineers. Usama Syed, the CEO, is a board-certified dermatologist trained at Imperial College London who has taught at Mount Sinai in New York, built an AI skincare database, and gathered more than 300,000 social media followers along the way. Hassan Mirza, the chief product officer, trained convolutional neural networks to sharpen ultrasound imaging at University College London while moonlighting as a front-end developer. Nabil Belahouane, the chief technology officer, was one of a handful of site-reliability engineers at the fintech GoCardless and a founding engineer at Documatic. Riyan Moussaoui rounds out the founding team.
Between them, the founders count more than 10,000 real patient encounters. That matters for a specific reason: it is easy to build a slick intake bot if you have never watched a patient bury the important symptom in the last thirty seconds of a visit. The people writing FutureClinic's medical reasoning have sat on the other side of that clipboard.
Syed's path is a small case study in how modern healthcare founders get made. A dermatologist who spent years explaining skin conditions to a large online audience learns two things most clinicians never do: how patients actually phrase their worries, and how much unmet demand sits just outside the traditional appointment system. Pair that with engineers who have shipped production software at fintech scale, and the team ends up unusually equipped to build the boring, load-bearing parts - billing, licensing, uptime - that sink most health startups long before the clinical model gets a fair test.
201 doctors, 57 specialties, 29 states
The specialties on offer run well past the usual telehealth staples - family medicine, dermatology, internal medicine, pediatrics, cardiology, and dozens more. For the patient, the experience is a messaging thread with a named human. Most doctors reply within a few hours and almost always the same day; occasionally a plan takes up to two days, the company notes, precisely because a real person is writing it.
The asynchronous format is doing quiet work here too. Because the conversation happens over messages rather than a scheduled video call, a patient can add a photo of a rash at 11 p.m. and the doctor can answer between other cases the next morning, without either party negotiating a calendar. It removes the single most annoying friction in outpatient care - the appointment - and replaces it with something closer to how people already talk to everyone else in their lives. The tradeoff is that anything genuinely urgent belongs in an emergency room, not a chat thread, and the platform routes only cases suited to virtual care.
Two steps, and no receptionist
Charge the doctor nothing, and only win when they do
The economics are where FutureClinic starts to look less like an app and more like infrastructure. A doctor pays nothing to join and no monthly fee. FutureClinic keeps a share of what the doctor charges, and in exchange runs the parts of a medical practice that are tedious and expensive to build alone: the electronic health record, payments, malpractice coverage, multi-state licensing, and a support desk. The doctor keeps the majority of every payment, sent weekly.
Patients pay in one of two ways. A one-off consultation typically runs $25 to $100, with some visits starting as low as $5. Many doctors also offer a monthly subscription, usually $50 to $100, that covers as many consultations as the patient needs that month. Prices vary by doctor and specialty, so the company openly suggests shopping around - a strange thing for a marketplace to say, and a revealing one.
One design choice does more work than the rest. On FutureClinic, a doctor earns the same whether or not they prescribe a drug or order a test. Unhook the advice from the sale, and the advice becomes something a patient can trust. That single decision is arguably the whole product.
Where it sits on a crowded shelf
Virtual care is not an empty market. Hims & Hers and Ro built large direct-to-consumer businesses, Teladoc scaled telehealth to the enterprise, Amazon launched Amazon Clinic, and K Health and Sesame each staked out a lane. Much of that field is organized around volume: a patient meets whichever clinician is next in the queue, and the model quietly resembles an assembly line with a nicer waiting-room app.
FutureClinic's wager is that people want the opposite - the same personal doctor over time, minus the appointment. Where most platforms own the patient and rent the physician's time, FutureClinic tries to hand the physician their own clinic and take a cut. It is closer to Shopify for doctors than to a call center for medicine.
That framing has a knock-on effect on supply. A dermatologist with a following, a primary-care doctor who wants to work from anywhere, or a specialist looking to see patients across state lines all get the same thing: a way to practice without opening an office, hiring front-desk staff, or wrestling a licensing bureaucracy in nine different states. For clinicians weighing burnout against building something of their own, that is a real offer, and it is how a two-sided platform seeds its supply side before the demand side fully arrives.
Illustrative positioning, not market share. Bars reflect where each model concentrates its design, based on public descriptions.
Proof of a human as a feature
There is something worth noticing in the timing. As AI-generated medical advice becomes cheap and everywhere, FutureClinic decided the scarce, valuable thing would be evidence that a real physician looked at your case and put their name on it. The Human Signature is that evidence, delivered as a face and a voice. The company that automated the most around the doctor also made the doctor the point.
FutureClinic is early. It has a small team, a YC stamp from Fall 2024, and the long, unglamorous road of state-by-state licensing and clinical trust ahead of it. But the shape of the bet is clear, and it is a coherent one: give the AI the paperwork, give the doctor the decision, and let a single physician finally scale without becoming a queue. Whether patients reward that is the experiment now running, one signed plan at a time.