Factory Scales Cloud Backend to Tens of Millions of Daily Requests on Vercel

Factory is handling tens of millions of API requests per day on a single Next.js backend on Vercel. It operates API routes, middleware, webhook handlers, and log pipelines without a separate infrastructure team, keeping p95 response time under 350ms.

Factory integrates its customer-facing API into its Next.js app, using middleware to manage authentication and routing. Webhook handlers and log drain pipelines also run on the same platform, and the company didn't build separate infrastructure or add new vendors while scaling.

Only memory limits and function execution time were adjusted, and Fluid Compute keeps functions warm between requests, reducing latency from cold starts. As a result, infrastructure hasn't become a bottleneck while the platform grew to support everyone from individual developers to enterprises with tens of thousands of engineers.

Non-technical teams use Droids in the Factory Desktop App or CLI to build internal tools and customer-specific prototypes themselves and deploy them to Vercel. Droids autonomously handle configuration, builds, and deployment through the Vercel API, letting engineers focus on setting guardrails and reviewing critical decisions.

Key use cases include:

• Internal analytics dashboards connecting analytics databases and software tools
• Customer-tailored demos that solutions engineers build and deploy within minutes
• Custom utilities for one-off internal tasks

This workflow enables dozens of deployments per day across teams and helps separate product development staff from internal tooling support work.