# How Superset built the IDE for AI agents on Vercel

DevFeed: [How Superset built the IDE for AI agents on Vercel](<https://devfeed.tech/articles/how-superset-built-the-ide-for-ai-agents-on-vercel-750.md>)

Original publisher: [Read original article](<https://vercel.com/blog/how-superset-built-the-ide-for-ai-agents-on-vercel>)

Author: Rini Vasan

Published: 2026-05-10T14:00:00Z

Content type: article

Language: en

Sources: [Vercel News](<https://devfeed.tech/sources/vercel-news.md>)

Topics: [AI-assisted coding](<https://devfeed.tech/topics/ai-assisted-coding.md>), [Developer Tools](<https://devfeed.tech/topics/developer-tools.md>), [ci](<https://devfeed.tech/topics/ci.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [ci](<https://devfeed.tech/tags/ci.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [coding](<https://devfeed.tech/tags/coding.md>), [deployment](<https://devfeed.tech/tags/deployment.md>), [ide](<https://devfeed.tech/tags/ide.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [next-js](<https://devfeed.tech/tags/next-js.md>), [platform-engineering](<https://devfeed.tech/tags/platform-engineering.md>), [vercel](<https://devfeed.tech/tags/vercel.md>)

## AI overview

Superset built an IDE for coordinating up to 10 coding agents in parallel across isolated workspaces and branches. The article explains how automatic preview deployments and parallel infrastructure on Vercel support that workflow.

## Source excerpt

Superset on Vercel 1,000-1,400 deployments per week ~600 preview deployments per day ~30 second average build time 57-64% week-over-week DAU growth Software development with AI started as a single engineer chatting with a single agent about a local repo. Today, developers direct fleets of agents in the cloud, but traditional tools were built for the old shape of the job: IDEs, terminals, and review systems designed for one developer moving one ticket at a time. Co-founders Kiet Ho, Satya Patel, and Avi Peltz, all former CTOs at YC-backed companies, built Superset as the IDE for multi-agent development. It runs up to 10 coding agents in parallel, each in its own isolated workspace. Developers use it to direct teams of agents generating code across multiple branches simultaneously. Running a team of agents in parallel changes what the platform underneath has to do. The product Superset offers its users only feels parallel because nothing on the platform forces the work to wait. If any layer slows down, even briefly, the parallelism on top collapses with it. Parallel agents need parallel infrastructure This workflow has a dependency that's invisible from the product surface. Every agent thread needs its own isolated environment, every branch needs a live URL, and every change needs a safe place to run. Without instant provisioning, parallel agents stop being parallel. CI pipelines have to be configured per branch, preview environments have to be managed by hand, and deploys back up behind one another. For a team running a dozen agents at once, that serialization is what breaks the product. Twelve workflows collapse into one queue, and a task that should take minutes takes hours. The developer is back to waiting, which is the exact problem Superset exists to solve. Six Next.js projects, no platform team Vercel was the default choice from the start, as all three founders had built on it at previous companies. From day one, Superset ran six Next.js projects on Vercel: the