The founders interviewed more than a thousand patients before writing a single line of code. Maybe that's why it seems to be working.
Juno, a San Francisco startup that emerged from Y Combinator, has built something that sounds almost too simple: an AI you can talk to about your chronic illness. Daily. Repeatedly. Without judgment or the ticking clock of a doctor's appointment.
The company claims 80,000 patients are using it globally as of May 2026—substantial for an app that launched last October—and says it's generating $50,000 in monthly recurring revenue. Those are the kinds of numbers that make investors look twice, especially when the team consists of exactly two people.
Marshall Gould and Isaac Tolley aren't your typical Silicon Valley dropouts building in a dorm room. Gould has a master's in genomic medicine from Oxford. Tolley studied molecular biology at University College London and spent time doing neuroscience research. They conducted those 1,000-plus patient interviews before launch, drawing on research from their respective universities, according to their Y Combinator profile. The diligence shows.
What they've built targets a sprawling, often overlooked population: people living with chronic conditions that mainstream medicine hasn't quite figured out. Fibromyalgia. Postural orthostatic tachycardia syndrome. Endometriosis. Long COVID. ME/CFS. The kind of illnesses that don't announce themselves in bloodwork, that fluctuate wildly day to day, and that leave patients trying to explain the inexplicable during a hurried office visit.
The Daily Grind of Being Sick
At its core, Juno is a mobile app—running on Anthropic's Claude, with voice infrastructure from LiveKit and text-to-speech services like ElevenLabs and Fish Audio. But what it does feels less technical than that stack suggests. Patients talk to it. They describe how they're feeling, what they ate, whether the fatigue is the heavy kind or the wired-but-exhausted kind. The AI responds conversationally, tracks patterns, and builds what Juno calls a "longitudinal health profile."
Then it generates PDF reports. Not the kind patients ignore, ideally, but the kind they can hand to a doctor and say: Here. This is what the last three months actually looked like.
For someone managing POTS or dysautonomia, where symptoms shift unpredictably and defy easy description, this kind of continuous documentation might matter more than another generic symptom tracker. Juno's design incorporates concepts familiar within the chronic illness community—spoon theory, pacing strategies, the vocabulary of invisible disability. Details that suggest the founders actually listened during those interviews.
The app supports chronic illnesses including fibromyalgia, POTS/dysautonomia, ME/CFS, endometriosis, autoimmune diseases, arthritis, migraines, IBS, EDS/hypermobility, MS, and Long COVID. It's an attempt to serve a population that's vast but fragmented, people who often feel the healthcare system wasn't built for them. (Because, frankly, it wasn't.)
The Numbers, With Caveats
Growth metrics from early-stage startups can be slippery, and Juno's trajectory offers a case study. The company says 80,000 patients are using the platform globally as of May 2026. In February, the founders mentioned 20,000 downloads and 400 five-star reviews within three months of launch. By April, a LinkedIn post from Tolley claimed 50,000 users and two million conversations logged.
That's rapid growth. Whether those users are sticking around—and paying—is harder to verify from the outside.
The subscription model appears to be working, at least if the $50,000 monthly recurring revenue figure holds. That metric, posted to both Product Hunt and the company's YC profile, suggests several thousand paying customers. Juno offers weekly and monthly plans with free trials, though some Reddit users have complained about confusion around when paid tiers actually begin. Early-stage pricing friction, in other words. Standard stuff.
Funding is murkier. According to a March profile in Forbes, Juno has raised $575,000. The discrepancy with other figures isn't huge—possibly a matter of timing in how tranches closed—but Forbes typically gets these details right.
The company entered Y Combinator's Spring 2026 batch. YC's standard deal these days is $500,000 for around 7% equity, though how much of Juno's total raise came directly from the accelerator versus other angels or early backers remains unclear. The founders haven't broken that out publicly.
Privacy, Disclaimers, and the Usual Concerns
Juno leans into voice. Hard. Patients can speak to the app throughout the day rather than laboriously typing symptoms into fields. For someone dealing with brain fog—a hallmark of many chronic conditions—this isn't just a nice-to-have. It's the difference between logging nothing and logging everything.
But voice means data, and health data especially. Juno's privacy policy from March states that health information and conversations aren't sold, though user data does flow to Anthropic to generate responses. That's inherent to how the system works, but it's worth noting for anyone uneasy about third parties touching their medical details.
Standard disclaimers apply: Juno isn't a licensed medical professional, shouldn't be used for emergencies, isn't a substitute for actual clinical care. Boilerplate, but necessary.
A Crowded Space, Sort Of
Juno isn't operating in a vacuum. Hello Heart launched an AI assistant called Nia for cardiovascular monitoring last November. K Health has built AI-powered tools aimed at providers rather than patients. Apps like Bearable and Manage My Pain have offered symptom tracking for years, though without a conversational AI layer.
What Juno seems to be betting on—and this is perhaps its real differentiator—is empathy at scale. Not just logging, but companionship. An AI that learns your patterns, remembers your triggers, and doesn't make you feel like you're complaining too much. For people managing conditions that doctors struggle to understand, let alone treat effectively, that might be worth paying for.
The addressable market is enormous, if a little fuzzy. CDC data shows three in four U.S. adults have at least one chronic condition; more than half have two or more. Globally, the World Health Organization reports that noncommunicable diseases account for the majority of deaths. Juno's claim of a "billion-plus" potential users is ambitious but not absurd.
What's Next, Maybe
The company has hinted at wearables integration. Reddit posts from the founders in February mentioned it was "coming soon," which would allow Juno to pull in data from fitness trackers and other devices. That would make the longitudinal profile more robust—sleep quality, heart rate variability, steps, all correlated with self-reported symptoms.
For now, Juno is available on iOS globally and reportedly on Android, though the Google Play Store listing proved elusive during research. APK aggregators show the app exists under the package name com.junochat.app, but a direct Play Store link wasn't easy to surface. Growing pains, or regional rollout quirks. Hard to say.
User feedback on Reddit cuts both ways. Some chronic illness patients describe the app as genuinely helpful, a rare tool that understands their reality. Others have griped about paywalls and unclear pricing structures. That tension is typical for consumer health products still hunting for product-market fit.
Juno's central wager is that AI can handle the tedious, grinding work of symptom tracking—the work that's exhausting for patients and often invisible to doctors constrained by fifteen-minute appointment slots. Whether that wager pays off depends on two things: whether people with chronic illness find enough sustained value to keep subscribing, and whether the AI can surface insights that actually improve health outcomes, not just documentation.
The early traction suggests the first part might be happening. The second part? That's the harder question, and the one that will determine whether Juno becomes a fixture in chronic care or just another health app that sounded promising in the pitch deck.
