We fetched exonic.ai seven ways on November 10, 2026: the homepage with and without www, /team, /media/cre, /studio, /sitemap.xml and a made-up address. Every one returned the same thing to a reader that does not run JavaScript: the line "Exonic – AI for adaptive medicine." The headline visitors see after scripts run ("Imagine if medicine could adapt to disease"), the team, the media page and the science never reach anything that reads raw HTML. That includes many AI crawlers, link previews in Slack, LinkedIn and email, and the scraping tools behind investor and partner databases. /sitemap.xml returns the same page instead of XML, so the site scan found 0 URLs. Made-up addresses also get the homepage instead of a "not found" page. Meanwhile directories fill the gap with an older "crowdsourcing drug discovery" story and conflicting facts. For a company whose buyers check the science before they reply to an email, the site is invisible exactly where it is being checked.
First action
Fetch exonic.ai/team with JavaScript off, confirm only the title line comes back, and list the five pages to prerender.
How to measure it
Indexed pages showing real content
- Now
- –
- Target
- 5 of 5 core pages, plus a correct preview on LinkedIn and X
- By
- 30 days
Steps
Confirm the gap
Fetch each page with JavaScript off and record what comes back; save it as the baseline.
Founder and CEO 1 hour
Write the page text
One headline, one paragraph and a meta description each for home, team, media, studio and contact.
Founder and CEO 1 day
Prerender
Switch those routes to static generation or server rendering so the text is in the first HTML response.
Contract web developer 2-3 days
Real sitemap and 404s
Serve an XML sitemap of the real pages and a 404 status for unknown paths.
Contract web developer half a day
Add markup and previews
Organization schema with links to YC, LinkedIn and X; Open Graph title, description and image on every page.
Contract web developer half a day
Submit and verify
Submit the sitemap in Search Console and request indexing of the five pages.
Founder and CEO 1 hour
Check like a partner would
Share links in LinkedIn and Slack, and ask two AI assistants "what does Exonic do?" a fortnight later.
Founder and CEO 2 weeks
Why it works
Prerender or statically export the pages that matter: home, team, media, studio and a new contact page. Each should have its real headline, a one-paragraph description of what Exonic does, a meta description, Open Graph tags and a canonical URL in the HTML the server sends. Generate a real sitemap.xml listing those pages and return a 404 status for addresses that don't exist. Add Organization markup that links to the YC profile, LinkedIn and X accounts. Then check the result the way a crawler would: fetch each page with JavaScript off, paste the URL into a LinkedIn post preview, and ask two AI assistants what Exonic does.
Partners, investors and candidates now research a company through search, AI answers and link previews before any call, and all of those read the HTML first. Prerendering five pages is a few days of work on a Vercel-hosted site. It turns the site from one line into the main, citable source on Exonic, and it fixes how the site looks when shared at no extra cost.
Ready to use
ChecklistCrawlable-site checklist
[ ] Home, team, media, studio and contact return their headline and a paragraph with JavaScript off [ ] Unique <title> and meta description per page [ ] Open Graph title, description and image per page [ ] Canonical URL on every page (www or apex, one only) [ ] /sitemap.xml returns XML listing the real pages [ ] Unknown paths return HTTP 404 [ ] Organization schema with sameAs links to the YC profile, LinkedIn and X [ ] Sitemap submitted in Search Console [ ] LinkedIn and Slack previews checked
Risks and how to handle them
Prerendering breaks interactive parts of the site
Prerender only the text pages and leave Studio as a client-side app behind its own route.
The message is still in flux
Write the page text after agreeing one description (idea 4); it can be edited later in minutes.
Evidence
- Your websiteEvery address fetched on November 10, 2026 (the homepage with and without www, /team, /media/cre, /studio, /sitemap.xml and a made-up path) returned the same thing to a reader that does not run JavaScript: the single line "Exonic – AI for adaptive medicine." robots.txt allows all crawlers.Source
- Your websiteSite scan (November 10, 2026): three sitemap files read but 0 page URLs listed; 0 blog posts; no contact, about, careers or pricing page found; the homepage links to /team, a media page at /media/cre and one science@ email address.Source
- PageSpeedMobile performance 25/100 (desktop 64). In Google's phone lab test the main content (the headline "Imagine if medicine could adapt to disease") appears after 14.4 s, with 27.6 s of blocking time; the homepage downloads 13.8 MB, including 2.7 MB of JavaScript and 5 KB of images. Largest fixes: cache lifetimes (about 11.8 s, 2.6 MB) and unused JavaScript (about 8.7 s, 2.0 MB). Real Chrome visitors (75th percentile) see the main content in 2.3 s, which passes, but responsiveness (INP) is 389 ms, above the 200 ms "good" line; layout shift is 0. Seven third-party services load, including Google Tag Manager (186 KB) and Google ad tags.Source
- Directory profilesHealthcare Discovery's profile says nearly all public information about Exonic comes from directory listings and company statements. It found no paper or preprint for the HepG2 result and no NIH or NSF awards; funding appears as $500K (PitchBook, Tracxn, with different close dates) and as $1M elsewhere; some directories name founders that no other source confirms.Source