Marshall Gould wasn't supposed to build a startup. After finishing his master's in genomic medicine at Oxford, the expected path led toward academic fellowships or a comfortable perch at a pharmaceutical company. Instead, he teamed up with Isaac Tolley—a neuroscience researcher from University College London who'd spent time building medical chatbots for the NHS—and disappeared into development mode for eight months.
What emerged was Juno, an AI health assistant designed specifically for people navigating chronic illness. Eight months after launching, their two-person operation had attracted 90,000 patients worldwide and was pulling in $70,000 monthly recurring revenue.
The trajectory is unusual, even by Y Combinator standards.
Juno went live in October, according to the company's Y Combinator profile. That means Gould and Tolley moved from concept to a functioning product serving tens of thousands of users in under a year. More remarkable still: they accomplished this entirely through organic content. No paid user acquisition. No growth hacking playbook. Just content that somehow found its audience.
Starting With Conversations, Not Code
Before they wrote their first line of software, the founders did something increasingly rare in the move-fast-and-break-things culture of consumer tech. They talked to people. A lot of people.
Through their academic positions at Oxford and UCL, Gould and Tolley conducted over 1,000 patient interviews. They weren't trying to validate assumptions or check boxes for investors—they were documenting, with the methodical rigor of researchers, what people living with chronic conditions actually needed. Day to day. Appointment to appointment.
The product reflects that groundwork. Juno supports a specific constellation of conditions: fibromyalgia, POTS (postural orthostatic tachycarma syndrome), ME/CFS (myalgic encephalomyelitis, or chronic fatigue syndrome), endometriosis, lupus, Long COVID. These are illnesses defined by what's hard to see—invisible symptoms, erratic flare-ups, and a healthcare system that often treats them as afterthoughts.
The interface is conversational, accepting voice or text. Users log symptoms through natural dialogue. The app tracks patterns across energy levels, pain thresholds, and other markers, then generates PDFs formatted for medical appointments. It's explicitly not diagnostic—the terms of service make clear that Juno "is NOT a licensed medical professional"—but it positions itself as something adjacent: a tool that helps people articulate what's happening when language fails them.
The Numbers Tell Their Own Story
Growth arrived in waves, then accelerated.
By late spring, the founders were reporting 50,000 users and 2 million logged conversations. A few weeks later, according to a profile in Founderland, those figures had climbed to 80,000 users with $50,000 in monthly recurring revenue. Around the same time, the company's Y Combinator launch materials cited more than 9,000 five-star reviews.
As of early summer, Juno's Y Combinator profile listed 90,000-plus patients and $70,000 MRR. The iOS app holds a 4.8 rating across 8.3K reviews. On Android—where the app launched mid-March—it crossed 5,000 installs within about two months and carries a 4.76 rating based on approximately 640 reviews.
Do the math: that's revenue nearly doubling and the user base expanding by 40,000 in roughly six weeks. For a bootstrapped health app with two founders and no marketing budget, that's an eyebrow-raising pace.
Under the Hood
Technically, Juno is built on straightforward infrastructure. The conversational layer runs on Anthropic's Claude. Voice capabilities come from LiveKit, with speech-to-text handled by Deepgram. Text-to-speech duties are split among ElevenLabs, Fish Audio, and Cartesia. Analytics flow through PostHog and Mixpanel.
The privacy policy, last refreshed in March, notes that user messages are sent to Anthropic to generate responses—a detail that's surfaced in Reddit discussions, where some users have raised questions about data handling and subscription trial transparency.
Juno claims "HIPAA-aligned infrastructure" in its documentation. The terms are explicit about boundaries: this is not emergency care, and it's providing educational information, not medical diagnosis.
How Content Became Growth
Perhaps the most intriguing element of Juno's rise is how it happened. In their YC Launch post, the founders were direct: they scaled "primarily through our organic content."
No ad spend. No carefully orchestrated influencer campaigns. Just content that connected with a community accustomed to being overlooked by the broader wellness industry.
Chronic illness affects perhaps a billion people globally—a figure the company references often—but most consumer health tools skew toward fitness tracking or acute symptom management. Juno identified something adjacent to a market gap: the daily slog of managing conditions that are long-term, variable, and frequently misunderstood by medical professionals themselves.
The App Store reviews reflect that resonance. Users describe the app as "life-changing." Many report bringing Juno's insights to doctor's appointments with tangible results. On its launch materials, the company suggests that "users have used her insights to get earlier diagnoses," though peer-reviewed clinical evidence specific to Juno wasn't publicly available as of mid-2026.
The Business Behind the App
Juno monetizes through a tiered subscription model: weekly options range from $4.99 to $6.99, monthly from $9.99 to $19.99, with quarterly at $44.99 and annual subscriptions spanning $39.99 to $99.99. The structure includes promotional rates, visible in iOS in-app purchase listings.
Corporately, there are signs of a quiet migration. The founders initially incorporated SharedGenes Ltd in the UK in September 2025, registering under business codes for both software development and human health activities. But by mid-March, that entity filed to strike off with Companies House, with the first gazette notice appearing late in the month. Meanwhile, Android app metadata lists "SharedGenes, Inc." as the developer, and CB Insights places the company's headquarters in San Francisco.
The implication: the team shifted operations to the U.S., likely coinciding with their entry into Y Combinator's spring cohort.
CB Insights records show a $500,000 convertible note from YC, consistent with standard investment terms, landing around mid-May.
What This Becomes
Ninety thousand users eight months post-launch. Revenue climbing at something close to 40% month-over-month in recent periods. Juno has momentum, no question. What's less clear is whether this is a devoted niche or the foundation of something that scales beyond a specific community.
The founders are researchers by training, not serial entrepreneurs. Tolley built healthcare software for Britain's National Health Service. Gould published three scientific papers during his master's work. These are people who learned to ship consumer software quickly, not Silicon Valley operators pivoting into digital health because the sector looked hot.
Which might be precisely what this space needs—not another wellness app calibrated for the worried well, but a tool constructed by people who understand the underlying science and, perhaps more importantly, listened first.
Whether that translates into a venture-scale company remains an open question. The numbers, at least for now, suggest they're onto something. Or maybe they just found an underserved audience hungry for a tool that actually speaks their language. In health tech, sometimes that's the same thing.
