k
keynews.ai
keynews.ai
ToolsThe story, in brief

How Factory scaled its cloud backend to tens of millions of daily requests on Vercel

Factory scaled to tens of millions daily requests on a single Next.js backend—without a dedicated infrastructure team. Here's how they did it.

Reporting by Vercel Blog 2 min brief
Paper-cut illustration of a coral software window opening into a three-dimensional drafting space.
New tools for building and creating with AI.AI illustration by KeyNews
The KeyNews take

Why it matters

A vendor case study showing how an AI agent platform (Factory) uses its own product internally to ship faster, then publicly documents the infrastructure pattern. Relevant to practitioners building scalable AI/agent backends and those evaluating Vercel for production AI workloads.

The key facts

15 to know

  1. Tens of millions of backend API requests served daily on single Vercel deployment

  2. 350ms p95 response time achieved with Fluid compute and Next.js

  3. Factory's own 'Droids' (autonomous agents) handle internal tooling, builds, and deployments—dozens per day

  4. Non-technical teams (sales, solutions engineers) now self-serve internal tool deployment via agents

  5. Scaled to multi-surface platform (API, Desktop App, CLI) without standing up separate infrastructure

  6. Security: Vercel Observability + WAF deployed to handle botnet attacks and fraudulent account creation

  7. No dedicated infrastructure team required; only tuning memory limits and function duration

  8. Tens of millions of backend API requests daily on single Vercel backend

  9. p95 response time: 350ms

  10. Factory uses its own Droids agents to handle deployments, configuration, builds end-to-end

  11. Non-technical teams (account executives, solutions architects) now self-serve deploy via Droids instead of requesting engineering

  12. Dozens of internal deployments per day across team

  13. All infrastructure runs on Next.js (not separate backend services)

  14. WAF and Vercel Observability handle API security and bot protection

  15. No dedicated infrastructure team required

Go to the source

Vercel Blogvercel.com

Publisher excerpt: Factory on Vercel Non-technical teams were bottlenecked on engineering Droids hit Vercel's APIs to build and deploy autonomously Going self-serve exposed Factory's backend to a new category of threat Observability and WAF handle protection at the network level Tens of millions of backend API…
Read original report
Back to today's editionMore tools news

The wider picture

View all
Paper-cut illustration of a coral software window opening into a three-dimensional drafting space.
AI illustration by KeyNews
Tools01

OpenAI nabs key Patreon execs ahead of upcoming announcement

OpenAI is building a creator-focused product suite with deep domain expertise (Patreon's co-founder + product + engineering leads). This signals a major new revenue and engagement vector for ChatGPT — and a direct threat to Patreon's existing creator economy.

The Verge AI
Illustration of a transparent lens revealing connected networks across layers of paper.
AI illustration by KeyNews
Tools02

Nokia Open-Sources AnyJev: A Training-Free Layer That Turns Any Open LLM Into a Calibrated Decision Model

A practical, deployment-ready layer for a common production pattern — classification over generation — that practitioners can drop into existing LLM stacks immediately. No fine-tuning required.

MarkTechPost
Paper-cut illustration of a coral software window opening into a three-dimensional drafting space.
AI illustration by KeyNews
Tools03

SpeakON Ships a MagSafe AI Voice Button With Its Own Microphone

A hardware-first approach to voice AI workflow — MagSafe button with onboard mic addresses the real friction in voice-to-text-to-action. Relevant to practitioners building voice UX and to the broader consumer AI tooling wave.

MarkTechPost