You have working Python code. Tidepool gives it a URL, a database, auth, payments, and email — no Docker, no AWS, no config files. Tell your AI to deploy it here.
tp runtime, and deploys it. You don't learn a framework — your AI does.main.py imports from your other modules. Your code stays organized.The runtime your project needs to live on the internet. No pip install, no config, no SDKs.
@tp.route Handler routingRegister request handlers with @tp.route('/path'). Pattern params, GET/POST, return HTML or JSON.
tp.files Persistent storageRead, write, list, delete files. Your data persists across requests. CSVs, JSON, images — up to 50GB per pod.
tp.page Static pagesRegister static HTML pages with tp.page('/about', html). Served from memory, no handler call needed.
tp.http HTTP requestsCall any API from your handlers. 200 requests per 60s. Your existing requests calls map directly.
tp.email Send emailSend plain text or HTML emails. Alerts, digests, reports — no SMTP setup needed.
tp.auth Full auth systemLogin, signup, Google OAuth, password reset — all built in. One dict configures everything.
Full-featured web apps, deployed in an afternoon.
A Substack-style site with auth, paid subscriptions, email tiers, and an admin panel. Multi-page routing, Stripe billing, the works.
A password-protected dashboard that pulls data from your APIs, stores history, and emails weekly reports to your team.
A lightweight API that receives webhooks, processes data, stores results, and serves a status page. JSON in, HTML out.

Tell your AI to deploy to Tidepool. One command, one URL, done.
Quickstart