Hormonaly
Evidence OS for modern medicine using AI agents to synthesize clinical answers from PubMed studies.
Website: https://hormonaly.ai/
Cover Block
PUBLIC
| Field | Value |
|---|---|
| Name | Hormonaly |
| Tagline | Evidence OS for modern medicine using AI agents to synthesize clinical answers from PubMed studies |
| Business Model | SaaS |
| Industry | Healthtech |
| Technology | AI / Machine Learning |
| Founding Team | Solo founder (Fady Hannah-Shmouni, MD FRCPC) |
Links
PUBLIC
- Website: https://hormonaly.ai/
- Consumer-facing product (Anabol.ai): https://www.anabol.ai/
- LinkedIn (founder): https://www.linkedin.com/in/fady-hannah-shmouni/
- Instagram (founder/product promo): https://www.instagram.com/p/DV_tSw-kUfq/
Executive Summary
PUBLIC
Hormonaly is positioning itself as an evidence operating system for clinical medicine, using a multi-agent AI architecture to retrieve and synthesize peer-reviewed literature into structured clinical answers [Hormonaly.ai]. The pitch, as stated on the company site, is that 24 specialized agents return GRADE-rated synthesis from a corpus of more than 10,000 PubMed studies in under 100 milliseconds [Hormonaly.ai]. The first vertical built on the engine is Anabol.ai, an evidence platform covering steroids, peptides, SARMs, and regenerative medicine, with 240-plus compound profiles, dosing protocols, side-effect data, and drug interaction information [anabol.ai]. The company is led by Fady Hannah-Shmouni, MD FRCPC, an endocrinologist and metabolic geneticist who previously served as associate program director of the NIH Inter-Institutes Endocrinology and Metabolism Fellowship Program and as a principal investigator in the Stratakis Laboratory's section on endocrinology and genetics [Erdheim-Chester Disease bio]. He sits on the editorial board of Endocrine and Metabolic Science at Elsevier [Elsevier]. Funding stage, headquarters, capitalization, and revenue are not publicly disclosed in the sources reviewed for this report. The next 12 to 18 months will turn on whether the Anabol.ai consumer surface produces enough usage data and design partner interest to support an institutional fundraise around the underlying agentic engine, and whether a clinician-facing product follows the consumer launch.
Data Accuracy: GREEN -- Confirmed across Hormonaly.ai, anabol.ai, LinkedIn, Elsevier, and the Erdheim-Chester Disease foundation bio.
Taxonomy Snapshot
| Axis | Value |
|---|---|
| Business Model | SaaS / agentic evidence engine |
| Industry / Vertical | Healthtech, clinical decision support, performance and hormone therapeutics research |
| Technology Type | Multi-agent AI, retrieval and synthesis over biomedical literature |
| Founding Team | Solo clinician founder |
Company Overview
PUBLIC
Hormonaly presents itself publicly as the "evidence OS for modern medicine," a platform layer designed to compress the work of literature review into structured, GRADE-rated answers a clinician or researcher can act on [Hormonaly.ai]. The first commercial expression of that engine is Anabol.ai, described on its own site as an "evidence-based AI research platform for steroids, peptides, SARMs and regenerative medicine" with more than 240 compound profiles drawing on PubMed-sourced studies, side effects, drug interactions, and dosing protocols [anabol.ai]. The founder, Dr. Fady Hannah-Shmouni, has used his LinkedIn and podcast appearances to frame Anabol.ai as a way to bring "structure, clarity, and scientific rigor to one of the most misunderstood and rapidly evolving domains in healthcare, anabolic, hormone, and performance-based therapeutics" [LinkedIn].
The company's headquarters, date of incorporation, and legal entity are not stated in any of the public sources reviewed. The most defensible milestones on the public record are the live deployment of the Hormonaly.ai marketing site describing the 24-agent architecture, the launch announcement of Anabol.ai as the first vertical product powered by that engine, and the founder's ongoing podcast circuit (Mizter Rad Show, PHAMCAST, Leaders On Purpose) where he has discussed personalized medicine, the shift from "sickcare" to "healthcare," and the role of agentic AI in clinical research [Mizter Rad Show; PHAMCAST; Leaders On Purpose Podcast]. No funding rounds, accelerator participation, board composition, or institutional investors have been disclosed in the captured sources.
Data Accuracy: YELLOW -- Two primary product sites and a credible founder LinkedIn confirm product existence and positioning; corporate formation details and timeline remain unverified.
Product and Technology
MIXED
The core product, as described on the marketing site, is an agentic retrieval-and-synthesis layer over biomedical literature. Hormonaly's site states that "24 specialized agents" produce "GRADE-rated clinical answers from 10K+ PubMed studies in under 100ms" [Hormonaly.ai]. GRADE (Grading of Recommendations, Assessment, Development and Evaluations) is a recognized framework in evidence-based medicine, and the company's claim is that its agents apply that framework to retrieved literature rather than producing free-text summaries. The 100-millisecond latency claim and the 10,000-study corpus size are company-stated and are not independently benchmarked in the sources reviewed.
Anabol.ai is the first applied surface and the most concrete way to evaluate what the underlying engine produces. Its public site advertises "240+ compound profiles" spanning steroids, peptides, SARMs, and regenerative-medicine compounds, with structured fields for PubMed-cited studies, side effects, drug interactions, and dosing protocols [anabol.ai]. The vertical choice is interesting: it is a domain where high-quality, structured evidence is genuinely scarce on the open consumer web, where users are highly motivated to research, and where the regulatory status of the substances pushes information-seeking out of mainstream clinical channels. That creates a real wedge for an evidence-first product, while also raising obvious content-policy and liability questions any institutional investor will want answered.
The underlying technical stack (model providers, vector database, retrieval pipeline, evaluation harness) is not described publicly. Inferences about infrastructure cannot be made from job postings because no open roles surfaced from major ATS hosts during research. What is observable is that the product is a retrieval-augmented system over a defined biomedical corpus rather than a generic medical chatbot, and that the company is presenting an explicit evidence-grading layer as its differentiation rather than the model layer itself.
Data Accuracy: YELLOW -- Product claims are confirmed on company-controlled surfaces; latency, corpus size, and agent architecture are not independently verified.
Market Research and Opportunity
PUBLIC
Clinical evidence synthesis is one of the few healthcare AI use cases where the workflow pain is acknowledged by clinicians themselves and where the underlying data (PubMed, Cochrane, guideline bodies) is largely public, which lowers the data-acquisition barrier that has slowed other healthtech AI categories.
The addressable market splits into at least three distinct buyers. The first is the practicing clinician who currently relies on point-of-care references such as UpToDate, DynaMed, and increasingly general-purpose LLMs; the value proposition there is faster, citation-anchored answers during a patient encounter. The second is the medical affairs and HEOR function inside pharmaceutical and medical-device companies, where literature reviews and evidence dossiers are produced continuously and where willingness to pay per seat is materially higher. The third is the researcher and the informed consumer, which is the surface Anabol.ai is testing first. None of these segments has a dollar size cited in the sources captured for this report, so any TAM figure here would be speculative; the conservative read is that point-of-care reference is a multi-hundred-million-dollar global subscription category historically dominated by Wolters Kluwer's UpToDate and EBSCO's DynaMed, and that the medical affairs evidence-synthesis spend sits adjacent to it.
Demand drivers visible in the founder's own framing include the shift in physician behavior toward using LLMs at the point of care (acknowledged in his podcast appearances on personalized medicine and AI in healthcare) [Mizter Rad Show; Leaders On Purpose Podcast], and the continuing growth of PubMed itself as the canonical biomedical corpus. Adjacent and substitute markets include general-purpose models (ChatGPT, Claude, Gemini) used off-label for clinical questions, specialized clinical AI products such as OpenEvidence and Glass Health, and the incumbent reference databases. Regulatory exposure is real but bounded: an evidence-synthesis tool that returns cited literature and is positioned as decision support rather than as a diagnostic device sits outside the strictest FDA software-as-medical-device categories, but a consumer surface like Anabol.ai that touches scheduled and unapproved compounds will face platform, advertising, and content-moderation risk that a clinician-only product would not.
No third-party market sizing report was surfaced in the captured research, so a sizing chart with confirmed numbers is not rendered here. The analyst takeaway is that the category is real and the buyers are identifiable; the open question is which of the three buyer segments Hormonaly chooses to monetize first, and whether the consumer Anabol.ai surface is a wedge to that decision or a distraction from it.
Data Accuracy: ORANGE -- Category exists and is well-documented in industry coverage broadly, but no sized market report is cited in the captured sources for this specific company.
Competitive Landscape
MIXED
Hormonaly is entering a category where the incumbents are well-capitalized reference publishers and the challengers are a new generation of LLM-native clinical answer engines, with general-purpose AI as the most-used substitute.
The competitive map is best described in segments. In point-of-care reference, the dominant incumbents are UpToDate (Wolters Kluwer) and DynaMed (EBSCO), both of which are deeply integrated into hospital subscriptions and clinician workflow and are now layering generative answers on top of their curated corpora. In LLM-native clinical answer engines, OpenEvidence has become the most visible challenger in the United States, distributing free to physicians and monetizing through pharma-sponsored placements, while Glass Health has pursued a clinician-facing differential-diagnosis and plan-generation product. In the consumer and prosumer evidence segment that Anabol.ai targets, the competitive set is more fragmented: Examine.com is the long-running incumbent for supplement evidence reviews, while general-purpose LLMs absorb the long tail of self-directed research queries.
Where Hormonaly has a defensible edge today, on the evidence available, is twofold. The first is founder credibility: Dr. Hannah-Shmouni's NIH research background and his standing as an endocrinologist and metabolic geneticist [Erdheim-Chester Disease bio; ResearchGate; Elsevier] give the product a clinical authority that is hard to manufacture and that matters disproportionately in the hormone and performance-therapeutics niche where Anabol.ai is launching. The second is vertical focus: by starting with a domain where the existing reference products are weak and where user motivation is high, Hormonaly can accumulate domain-specific structured data (compound profiles, interactions, dosing) that a horizontal competitor would have to fund deliberately to replicate. Both edges are real but perishable: founder credibility does not scale to other verticals without team build-out, and vertical data moats erode quickly once a horizontal incumbent decides the vertical is worth covering.
Where the company is most exposed is distribution. OpenEvidence has built its growth on direct-to-physician acquisition with a free tier; UpToDate is already inside the hospital contract; ChatGPT is already on the clinician's phone. A small team without disclosed institutional capital will struggle to outspend any of these on either consumer or enterprise distribution, which makes the choice of go-to-market wedge the single most important strategic decision in front of the company. The most plausible 18-month scenario splits along that decision. Winner if Anabol.ai produces a measurable usage base and a defensible safety-and-content posture, because that creates both a consumer brand and a dataset that can be repackaged for clinician and pharma buyers. Loser if the company tries to compete head-on with OpenEvidence in general clinician answers without the capital to fund free distribution at scale.
Opportunity
PUBLIC
If Hormonaly executes, the prize is to become the default evidence layer that sits between the world's biomedical literature and any clinician, researcher, or pharma team that needs a citation-anchored answer.
The single largest outcome Hormonaly could plausibly become is the structured-evidence API for medicine: the system that other clinical applications, EHR vendors, pharma medical-affairs tools, and consumer health products call when they need a GRADE-rated answer with citations rather than a free-text generation. The ingredients for that outcome are visible in the public evidence. The corpus (PubMed) is open. The framework (GRADE) is recognized. The founder is a published researcher with NIH provenance and editorial-board standing [Erdheim-Chester Disease bio; Elsevier]. The first vertical (Anabol.ai) is a market test in a domain where the answer quality can be judged by knowledgeable users [anabol.ai]. None of that guarantees the outcome, but it makes the outcome reachable rather than aspirational.
Growth scenarios.
| Scenario | What happens | Catalyst | Why it's plausible |
|---|---|---|---|
| Consumer wedge to clinician product | Anabol.ai builds a usage base in performance and hormone therapeutics, then the same engine is repackaged for endocrinologists and primary-care physicians | A measurable Anabol.ai DAU base plus a clinician-facing pilot inside an endocrinology practice | The founder's specialty is endocrinology and his audience already overlaps with the Anabol.ai user [LinkedIn] |
| Pharma medical-affairs sale | Hormonaly licenses the evidence engine to one or more pharmaceutical medical-affairs teams for internal literature synthesis | A named design-partner contract with a mid-cap pharma | Medical affairs is the highest-willingness-to-pay buyer for structured evidence and the workflow pain is acknowledged across the industry |
| Embedded evidence API | The engine is integrated into other clinical software (EHR add-ons, decision-support tools, consumer health apps) on a usage basis | A first integration partner ships with Hormonaly citations visible in-product | The company already presents itself as an "OS" rather than a single application [Hormonaly.ai], which signals API intent |
What compounding looks like. The flywheel in evidence synthesis is data and trust rather than network effect in the social sense. Each query produces a feedback signal on which retrievals and which GRADE assessments were useful, which improves the next answer. Each new vertical (after performance therapeutics, plausibly endocrinology, then metabolic disease, then adjacent specialties) adds a structured dataset that the horizontal incumbents have to build from scratch to match. Each clinician or researcher who cites a Hormonaly-generated answer in their own work creates a credibility footprint that compounds the founder's existing publication record [Google Scholar; ResearchGate]. The early evidence that the flywheel is starting is thin (no usage metrics are public), but the architecture is consistent with one.
The size of the win. A credible public comparable is Wolters Kluwer's UpToDate, long understood to be a multi-hundred-million-dollar annual subscription business inside a publicly traded parent. A more recent private comparable is OpenEvidence, which has raised institutional capital at a valuation widely reported in the technology press on the strength of physician adoption rather than disclosed revenue. If Hormonaly captures even a vertical slice of the medical-affairs or specialist-clinician evidence market while keeping the consumer Anabol.ai surface as a brand and data engine, a venture-scale outcome in the high hundreds of millions of dollars is a coherent scenario rather than a fantasy (scenario, not a forecast). The downside-bounded version is a profitable specialty-focused SaaS for endocrinology and performance medicine; the upside-unbounded version is the embedded evidence API for the next decade of clinical software.
Data Accuracy: ORANGE -- Scenarios are analyst constructions grounded in confirmed product positioning and founder background; outcome ranges are explicitly labelled as scenarios rather than forecasts.
Sources
PUBLIC
[Hormonaly.ai] Hormonaly.ai - The Evidence OS | https://hormonaly.ai/
[anabol.ai] anabol.ai - Research compounds. Make smarter choices. | https://www.anabol.ai/
[LinkedIn] Fady Hannah-Shmouni, MD FRCPC - Hormonaly | https://www.linkedin.com/in/fady-hannah-shmouni/
[Elsevier] Fady Hannah-Shmouni, MD, DABIM, AHSCP, FRCPC - Editorial Board, Endocrine and Metabolic Science | https://www.journals.elsevier.com/endocrine-and-metabolic-science/editorial-board/fady-hannah-shmouni-md-dabim-ahscp-frcpc
[Erdheim-Chester Disease] Fady Hannah-Shmouni, MD, DABIM, FRCPC - Webinar Bio | https://www.erdheim-chester.org/wp-content/uploads/2021/02/Hannah-Shmouni-Fady-Webinar-Bio.pdf
[ResearchGate] Fady Hannah-Shmouni - Research profile | https://www.researchgate.net/profile/Fady-Hannah-Shmouni
[Google Scholar] Fady Hannah-Shmouni, MD, DABIM, AHSCP, FRCPC | https://scholar.google.ca/citations?hl=en&user=j_JtDx0AAAAJ
[Mizter Rad Show] Dr. Fady Hannah-Shmouni: Personalized Medicine and the Future of Aging | https://www.mizter-rad.com/episodes/44-aging-by-choice-dr-hannah-schmouni
[PHAMCAST] Dr. Fady Hannah-Shmouni (Part 2) | https://podcasts.apple.com/ca/podcast/dr-fady-hannah-shmouni-part-2/id1700393383?i=1000665000074
[PHAMCAST] Dr. Fady Hannah-Shmouni (Part 3) | https://open.spotify.com/episode/2UnZfmlKpc8BszIIVbE69u
[Leaders On Purpose Podcast] Episode 47 - Dr. Fady Hannah-Shmouni: Be The CEO Of Your Health | https://open.spotify.com/episode/1wDp16c30HK04bG3J8up8E
[Instagram] Fady Hannah-Shmouni, MD - Anabol.ai promo | https://www.instagram.com/p/DV_tSw-kUfq/
Articles about Hormonaly
- Hormonaly Wants 24 AI Agents Reading PubMed for Every Clinician's Hardest Question — Endocrinologist Fady Hannah-Shmouni is building an evidence engine that grades the literature before it answers, starting with hormone therapeutics.