Teams build many small internal tools (via AI, low-code, or scripts) but managing hosting, auth, and uptime for a dozen little apps is a pain nobody wants to deal with.
A platform that hosts arbitrary small web apps (generated by any tool) with built-in auth, monitoring, and zero-config deployment — like Retool's hosting but for any codebase.
Subscription per app or per seat, tiered by number of apps and usage
The pain is real but moderate. Teams DO struggle managing dozens of small apps, but many muddle through with existing PaaS or just run things on a shared VM. The HN comments ('I just want a quiet life') signal genuine frustration but not hair-on-fire urgency. Pain intensifies as the number of micro-apps grows past ~10.
TAM for internal tools infrastructure is large ($10B+), but the specific niche of 'managed hosting for arbitrary small internal apps' is narrow. Your addressable market is small-to-mid engineering teams (5-50 people) building 10-50 internal tools — probably a $500M-$1B SAM. Not tiny, but you need strong density in a niche before expanding.
Tricky. Teams already pay for Railway/Render/Fly.io for hosting. The premium you're charging is for the auth layer, app catalog, and 'zero config' wrapper. Many teams will compare your $X/app price to just spinning up another $5 Fly.io machine. You need to bundle enough value (SSO, monitoring, app portal) to justify 2-3x the raw hosting cost. Internal tools budgets exist but are often an afterthought.
A solo dev can build an MVP in 6-8 weeks by wrapping an existing container runtime (Fly.io Machines API, or Docker on a VPS) with an auth proxy (OAuth2 Proxy/Authelia), a simple app registry UI, and a deploy pipeline. The hard parts come later: multi-tenancy security, resource isolation, supporting arbitrary codebases reliably, and uptime guarantees. MVP is feasible; production-grade is hard.
No existing product nails the exact combo of: (1) host ANY codebase, (2) built-in auth/SSO, (3) app catalog for end users, (4) monitoring, (5) truly zero-config. Railway/Render are close on hosting but lack the internal-tools layer. Retool has the internal-tools layer but only for Retool apps. The gap is real and specific.
Excellent subscription fit. Once teams deploy 10-20 internal tools on your platform, switching cost is very high. Per-app or per-seat pricing scales naturally with usage. Infrastructure hosting is inherently recurring. Expansion revenue is built-in as teams build more tools.
- +Clear gap in the market — nobody combines arbitrary-codebase hosting with internal-tools features (auth, catalog, monitoring)
- +AI code generation is creating a tailwind of more small apps that need hosting, growing demand organically
- +Very high switching costs and strong recurring revenue once teams are onboarded
- +The 'quiet life' positioning resonates — developers hate being accidental DevOps for internal tools
- !Willingness to pay is unproven — teams may prefer stitching together Railway + Auth0 + Datadog for less
- !Security surface area is massive — hosting arbitrary user code with shared auth is a hard trust/isolation problem
- !PaaS incumbents (Railway, Render, Fly.io) could add an 'internal tools' layer and eat your lunch in a quarter
- !Support burden: 'any codebase' means infinite edge cases in build/deploy, which is expensive to support at small scale
Modern PaaS for deploying apps from Git repos or Docker containers with automatic builds, databases, and environment management. Supports multiple services per project.
Cloud PaaS offering web services, static sites, cron jobs, and managed databases with Git-based auto-deploy and free SSL.
Edge-first PaaS that runs Docker containers as micro-VMs globally, with fast deploys and built-in private networking between apps.
Low-code platform for building internal tools with drag-and-drop UI components connected to databases and APIs. Includes hosting and auth.
Open-source, self-hosted PaaS alternative to Heroku/Netlify. Deploy any app, database, or service on your own servers with a web UI.
Start narrow: support only Dockerized apps and static sites. Provide a CLI/GitHub-based deploy flow, an auth proxy (SSO via Google Workspace), a simple web portal listing all deployed apps, and basic health monitoring. Target teams already using AI tools (Cursor, Claude) to generate internal apps. Skip custom domains, complex scaling, and multi-cloud — just make deploy-and-share dead simple for 10-30 small apps on a single cluster.
Free for up to 3 apps (hook teams) -> $10/app/month for 4-20 apps (core tier) -> $200-500/month flat for unlimited apps + SSO + audit logs (team tier) -> Enterprise with SLAs, on-prem, compliance. Upsell on usage (compute, bandwidth) as apps get heavier.
8-12 weeks to MVP, 12-16 weeks to first paying customer. The long pole is not the tech — it's finding 5-10 design partners willing to migrate their internal tools onto an unproven platform. Expect 3-4 months to first revenue.
- “somebody else is looking after the hosting of the dozen little one-purpose applications”
- “it all just works”
- “ultimately I'm just after a quiet life”