Computer for Builders
The number of solo founders has doubled in the past decade. Cloud infrastructure and automation tools made it easier to launch software without a big team. But as businesses grow, so does the work of adding features, fixing bugs, and managing the books. That’s the work a powerful agent platform can take on.
Since launching Perplexity Computer in February, we’ve seen founders increasingly rely on the multi-model orchestrator to run the entire software development loop. When connected to the right tools, Computer can write code, deploy it, monitor performance, track payments, and report results.
Today we’re announcing Computer for Builders, a collection of tools for founders and small engineering teams. A builder can add a feature, ship it, watch it in production, track revenue, and get a growth report without toggling between tools. Computer orchestrates 15+ frontier models and picks the best one for each task.
Small teams get time back to continue innovating. Work that used to sit in the backlog waiting for bandwidth now gets shipped.
Lets builders keep the focus on innovation
Connects to the tools builders already run
Builders and small teams rely on a trusted set of tools to build and run their businesses. These are the ones at the center of the Computer for Builders workflow:
GitHub: The world’s developer platform where teams host code, collaborate on pull requests, and track issues. Explore repos and code, create and review pull requests, and search and manage issues right from chat.
Datadog: The unified observability and security platform lets the teams see inside any stack, any app, at any scale. Query metrics, search and analyze logs, and inspect distributed traces, dashboards, and alerts right from chat.
Stripe: Handles online payments, subscriptions, and billing through developer-friendly APIs. Look up customers, review payments and subscriptions, and inspect invoices, payouts, and fraud signals right from chat.
Supabase: An open-source backend on Postgres that ships with a database, authentication, storage, and real-time APIs. Run read queries against Supabase tables to explore production data, inspect users, manage storage buckets, and trigger edge functions right from chat.
Slack: Organizes team communication into channels and centralizes notifications from other tools. Search channels, read message history, send messages across a workspace right from chat, and more.
Computer also accesses content in Google Drive, Google Calendar, Gmail, and 400+ other files and tools through App Connectors.
With the tools for builders, founders and teams skip the manual work. Computer connects to the codebase, opens PRs, deploys, watches production, tracks payments, and sends a weekly growth report.
Ships code, tracks revenue, and more
Computer helps founders and small teams get products live and measure the impact using the tools they already use. It handles:
Shipping code end to end: Computer connects to the repo, writes the code, and opens a PR on GitHub for the user to review and merge.
Deploying and watching production: Computer deploys through Vercel and watches production in Datadog, both performance security together, from latency spikes to exposed vulnerabilities. If something breaks or looks like an attack, it flags what happened, where, and what shipped right before it did.
Tracking revenue and fraud: Computer watches Stripe for cancellations, failed payments, and fraud signals, and flags what needs attention.
Querying data: Computer queries Supabase and other databases, then turns the results into a readable report.
Running recurring reports: Computer runs a weekly growth report and sends it to Slack every Monday automatically.
Works across many different use cases
Here are a few examples of how Computer works end to end:
Fixing bugs: Computer reads new GitHub issues, clusters the top recurring bug, pulls the relevant Supabase data and GitHub code to reproduce it, opens a PR with a test, deploys via Vercel, and posts a “fixed in this release” note back to the Slack #customer-channel.
Setting up subscriptions: Computer sets up products, prices, and a checkout flow in Stripe, gates access behind Clerk, wires it into to Supabase/Next.js app via a GitHub PR, deploys on Supabase, and confirms the first live test charge
Building an AI answer feature: Computer reviews the product requirements, maps the repo structure in GitHub, opens a PR that wires the Perplexity Agent API into their Next.js/Supabase backend, deploys via Vercel, tracks token usage, cost, and latency with Datadog’s Agent Observability and meters usage for billing through Stripe.
Turning churn signals into retention actions: Computer connects Stripe cancellation events with Supabase behavioral data to find the drop-off pattern, ships a small in-app nudge via a GitHub PR and Vercel deploy, then tracks whether the saved-subscription rate improves in a weekly Slack report.
Available to all Computer users
Computer for Builders tools are available to Pro and Max subscribers who use Computer.
New to Perplexity? Try Computer for free.