Sample site built by our AI, not a customer site. Fernpost is a made-up business; our AI wrote every section, from the headline to the 4-question FAQ. Fernpost turns messy insurance and medical documents into structured data with confidence scores and human review built in. Deployed in your own cloud.
The first thing a visitor reads is "Read 400 page claim files in ninety seconds", with "In production since 2022" set above it as a small label. The line underneath does the explaining: "Fernpost pulls dates, codes, providers and amounts out of scanned faxes, handwritten forms and PDF soup, scores its own confidence and routes anything uncertain to a human." Two buttons follow, "Book a pilot" for people ready to act and "Read the docs" for people who want to look around first. The layout is a split hero, copy on one side and footage on the other, using looping stock footage instead of a static photo.
Directly below the hero, a strip of 4 short facts does the job most ai startup sites leave to vague words like "trusted" or "experienced": 96.8% (field level accuracy), 90 sec (avg 400 page file), 31 (production customers) and SOC 2 (Type II audited). Each one is something a customer can check or hold the business to.
The "Pilot first, then volume pricing" section lists 6 how you buy it: Four week pilot ($18,000), Standard volume plan (From $0.22/page), In your cloud deployment (From $9,500/mo), Custom field training (From $12,000), Human review queue (Included) and API and webhook access (Included). Prices run from $0.22 to $18,000, so a visitor can tell in seconds whether it fits the budget. A one-line note sets the terms: "Every engagement starts with a paid pilot on your real documents, not a canned demo set." Each item gets a sentence of its own, for example "We run 2,000 of your real files, publish a measured accuracy report and you keep the results either way."
Under the heading "Accuracy you can audit, not a number in a slide", the about section says: "Fernpost was started in 2022 by two engineers from a claims platform and a former utilization review nurse who spent eleven years reading these documents by hand. That last part matters, because knowing which fields are worth being uncertain about is most of the problem." It then lists 3 concrete points instead of a paragraph of values: every extracted value links back to the source page region for audit, sOC 2 Type II audited, with BAAs available for covered entities and accuracy measured on your documents during the pilot, published in full.
There are 3 reviews, each signed with a name and a context (Lauren M., VP Claims Operations; Raj S., Director of Analytics, TPA; Kevin D., CISO, regional health plan). None of them says "great service". They mention a specific moment, like "The pilot report was refreshingly blunt about which field types it struggled with. That honesty is why we signed."
The FAQ covers 4 questions: "What does 96.8 percent accuracy actually mean?", "What happens when the system is unsure?", "Do our documents leave our environment?" and "Why is the pilot paid?". The answers are direct. The first one reads: "It is field level exact match across our production customer base, measured against human reviewed ground truth, not a document level or character level figure which would look higher and mean less. Accuracy varies by field type and document quality, so the pilot measures it on your specific documents and we publish the breakdown by field."
The page closes on "Run it on your worst files". It sets expectations before anyone gets in touch: "Send us your ugliest sample set, the faxes and the handwriting. That is the honest test and it is the one we would rather be judged on." Hours are written out (Mon to Fri 8am to 6pm MT, enterprise support 24/7) and so is the area covered (United States, EU deployments available). The phone number also sits in the header on every scroll position, so nobody has to hunt for it.
The palette is three colors: a #f5f6f8 background, #111827 text and one accent (#1d4ed8) used only for buttons and labels. Headings are set in a sans-serif face and buttons have 6px rounded corners. The menu has 6 items (Services, Pricing, About, Reviews, FAQ and Contact), all on one page, so it scrolls instead of making people click through.
Describe your business in a sentence and the AI writes a site like this one around it. Start building.
This page is for inspiration: one real example, taken apart. For the pages a ai startup site needs, what it costs and what the builder can and cannot do, see the ai startup website builder page.
Free plan: up to 3 template sites on a we.inc address, with no AI and no custom domain. AI building, like this example, starts at $20/mo, and custom domains are on paid plans. The AI builds request and contact forms; payments go through your own Stripe account.
We.Inc is an AI-powered website builder you can resell under your own brand. Launch a branded client dashboard, bill on Stripe Connect, and deliver AI-generated websites in minutes. White-label plans start at $99 a month for 25 client sites, with a 7-day free trial and no per-site fees.