Clara, a Y Combinator-backed AI primary care startup, is hiring a founding full-stack growth engineer to bring its AI doctors to market. The job posting on YC’s job board is the most revealing document about the state of AI healthcare that has crossed this newsroom in months. It is not a research paper. It is not a clinical trial. It is a job description, and it says more about how AI doctors will actually reach patients than any benchmark or model card.
Clara was founded in 2025, is in YC’s P26 batch, has a team of nine, and has raised $12 million in pre-seed funding from Y Combinator and top investors. The founders previously built Circle Medical to over $100 million in annual revenue, and founding executives came from Hims and Hers, GoodRx, and Teladoc. The company describes itself as “an AI-powered medical practice” where human medical providers review and approve every medical decision. The mission, as stated in the posting, is to “add one billion years of healthspan to humanity.”
The salary range is $130K to $210K with 0.10% to 0.40% equity. The role is in San Francisco, five days a week in person. Visa sponsorship is available for TN, OPT, and H1B transfers.
Here is what makes this posting remarkable. The job is not for a clinician. It is not for a machine learning researcher. It is for a growth engineer who owns the entire funnel, from the first ad impression through onboarding, first chat, subscription, and long-term retention. The responsibilities read like a playbook for consumer software: landing pages, onboarding funnels, experimentation, attribution, lifecycle messaging, referral loops, and CRM surfaces across email, SMS, and in-product notifications.
This is the story of AI healthcare in 2026. The hard part was never the model. The hard part is distribution, trust, and habit formation in a regulated industry where every claim must be medically accurate and HIPAA-safe.
The growth engineer as the clinical gatekeeper
The posting explicitly states that the growth engineer will “partner closely with our clinical and compliance teams to make sure every page, claim and experiment is medically accurate and HIPAA-safe.” That sentence is doing a lot of work. In traditional healthcare, the clinical team owns medical claims and the marketing team owns conversion. Clara is collapsing those roles into a single founding growth engineer who writes the code, ships the pages, wires the events, runs the tests, and reads the results.
This is a genuinely new organizational form. The growth engineer is not just responsible for acquisition metrics. They are responsible for the medical accuracy of every claim on every landing page. That is a profound shift in how healthcare marketing works. The person who decides whether a landing page says “treat your hypertension with Clara” is also the person who decides which experiment to run next. The clinical team reviews, but the growth engineer builds.
The posting also reveals the technical stack in unusual detail. The growth stack includes Amplitude with session replay and Experiment, Customer.io, Stripe, and Google, Meta, and Reddit Ads APIs. The product runs on React 19 with TypeScript, a Next.js and Payload CMS marketing site on Vercel, a Python and Django backend on AWS, and “frontier LLMs with proprietary prompts and tool calls.” This is a mature, opinionated stack for a nine-person company. The mention of Reddit Ads specifically suggests they are already testing community-driven acquisition channels, which is an interesting signal for a healthcare product.
What this means for the AI economy
The take here is not that Clara is special. The take is that Clara represents the maturation of AI healthcare as a category. The first wave of AI healthcare startups sold the model. The second wave is selling the practice. Clara is explicitly building “the autonomous medical practice of the future,” with human providers reviewing every decision. That is a regulatory posture as much as a product posture.
The job posting signals that the bottleneck for AI healthcare is no longer model capability. Frontier LLMs are a commodity input, mentioned almost in passing. The bottleneck is the growth engine: the ability to acquire patients at a cost that makes the unit economics work, to onboard them quickly, to get them to first value fast, and to keep them engaged over the long term. That is the same playbook as any consumer subscription business, except the stakes are health outcomes and the regulatory surface is enormous.
The phrase “Turn patient behavior signals into personalized in-product prompts, with clinical review in the loop” is worth pausing on. Clara is building a system where the growth engineer designs prompts that nudge patients toward better health behaviors, and every prompt is clinically reviewed. This is a new category of software: clinically-reviewed behavioral nudges at scale. The growth engineer is not just optimizing for retention. They are optimizing for health outcomes, which is a much harder objective function.
The uncomfortable truth about AI doctors
There is an uncomfortable truth buried in this posting. The mission is “one billion years of healthspan,” but the job is fundamentally about subscription retention. The growth engineer owns “subscription and long-term retention.” They own “reactivation and re-engagement.” They own “the logic that decides who hears what, when.”
This is the reality of AI healthcare in 2026. AI doctors will win or lose on the same metrics as any direct-to-consumer healthcare company: acquisition cost, activation rate, retention curve, and lifetime value. The clinical outcomes matter, but they matter through the lens of whether patients keep paying. The posting does not hide this. It embraces it. The growth engineer is the person who makes the mission financially viable.
The posting also reveals a specific bet about how AI healthcare will acquire patients. The first responsibility listed is “Ship programmatic and SEO-driven page systems for condition, medication and geo intent.” That is a search-driven acquisition strategy. Clara is betting that patients will search for their condition, their medication, or their location, and Clara will have a page that answers that search and converts it into a signup. This is the same playbook that Hims and Hers and Roman used to build billion-dollar companies, applied to an AI-native practice.
What to watch
The most interesting signal in this posting is the requirement for “12+ months in a dedicated growth engineering role” and “experience scoping and prioritizing initiatives using RICE or a similar framework, and defending the call.” Clara is not looking for a generalist. They are looking for someone who has done growth engineering as a discipline, who knows how to run experiments with guardrails, and who can defend prioritization decisions. That is a mature hiring pattern for a company that is nine people.
The nice-to-haves include “comfortable using LLMs as a production tool for content, creative and internal tooling.” That is a quiet acknowledgment that the growth engineer will use AI to generate the very content that acquires patients. The growth engine will be powered by the same frontier LLMs that power the clinical product. The acquisition content and the clinical product will be built on the same substrate.
Clara’s job posting is a window into the next phase of AI healthcare. The models are good enough. The regulatory posture is defined. The remaining question is whether a nine-person team in San Francisco can build a growth engine that acquires patients at scale, keeps them engaged, and does it all under HIPAA and clinical review. The founding growth engineer will answer that question, and the answer will determine whether AI doctors are a real category or a demo.
The posting closes with an interview process that includes a four-hour onsite with system design and code pairing. The first question in the intro call is about Clara’s mission. The last question, implicitly, is whether the candidate can defend their RICE scores against a clinical team that outranks them.
Tessera will be watching whether Clara can turn the growth-engineer role into a repeatable acquisition machine, and whether the “clinically-reviewed prompt” becomes a new standard for AI healthcare engagement.