Skip to content
How I AIHow I AI

How Stripe deploys 1,300 AI-written PRs per week

Steve Kaliski is a software engineer at Stripe who has spent the past six and a half years building developer tools and payment infrastructure. He’s part of the team that created “minions”—Stripe’s internal AI coding agents, which now ship approximately 1,300 pull requests per week with minimal human intervention beyond code review. In this episode, Steve demonstrates how Stripe engineers activate development work from Slack and leverage cloud-based development environments for parallel agent workflows, and demos machine-to-machine payments where AI agents transact autonomously with third-party services. *What you’ll learn:* 1. How Stripe’s “minions” write 1,300 pull requests per week with minimal human intervention 2. Why a good developer experience for humans creates better outcomes for AI agents 3. The critical role of cloud development environments in unlocking AI-powered engineering velocity 4. The machine payment protocol that lets AI agents spend money to accomplish tasks 5. The code review strategy for handling thousands of agent-written PRs 6. Why non-engineers at Stripe are starting to use minions to ship code 7. The future of software businesses built primarily for agent consumers *Brought to you by:* Optimizely—Your AI agent orchestration platform for marketing and digital teams: https://www.optimizely.com/howIAI Rippling—Stop wasting time on admin tasks, build your startup faster: https://rippling.com/howiai *In this episode, we cover:* (00:00) Introduction to Steve (02:39) Stripe’s minions and their effect on Stripe as a whole (04:42) Why activation energy matters more than execution (05:44) What is a minion? The technical architecture (06:52) Demo: Activating a minion from Slack with an emoji (09:04) Why good developer experience benefits both humans and agents (11:22) Walking through the agent loop and system prompts (13:42) Why Stripe chose Goose as their agent harness (16:00) The role of Stripe’s developer productivity team (17:15) Why cloud environments unlock multi-threaded AI engineering (21:14) One-shot prompting: from Slack to shipped PR (22:04) How Stripe handles code review for 1,300 AI-written PRs weekly (23:44) Non-engineers using minions across the company (24:53) Demo: Planning a birthday party with Claude and machine payments (32:15) Quick recap (35:08) The future of ephemeral, API-first businesses for agents (36:36) Lightning round and final thoughts *Detailed workflow walkthroughs from this episode:* • How Stripe's AI 'Minions' Ship 1,300 PRs Weekly from a Slack Emoji: https://www.chatprd.ai/how-i-ai/stripes-ai-minions-ship-1300-prs-weekly-from-a-slack-emoji • How to Build an Autonomous AI Agent That Pays for Services to Complete Tasks: https://www.chatprd.ai/how-i-ai/workflows/how-to-build-an-autonomous-ai-agent-that-pays-for-services-to-complete-tasks • How to Automate Code Generation from a Slack Message into a Pull Request: https://www.chatprd.ai/how-i-ai/workflows/how-to-automate-code-generation-from-a-slack-message-into-a-pull-request *Tools referenced:* • Goose (AI agent harness): https://github.com/block/goose • Claude Code: https://claude.ai/code • Cursor: https://cursor.sh/ • VS Code: https://code.visualstudio.com/ • Slack: https://slack.com/ • Browserbase: https://browserbase.com/ • Parallel AI: https://www.parallel.ai/ • PostalForm: https://postalform.com/ • Stripe Climate: https://stripe.com/climate *Other references:* • Stripe machine payments: https://docs.stripe.com/payments/machine • Blue-Green Deployment: https://martinfowler.com/bliki/BlueGreenDeployment.html • Git worktrees: https://git-scm.com/docs/git-worktree *Where to find Steve Kaliski:* Twitter: https://twitter.com/stevekaliski LinkedIn: https://www.linkedin.com/in/steve-kaliski-079a7710/ *Where to find Claire Vo:* ChatPRD: https://www.chatprd.ai/ Website: https://clairevo.com/ LinkedIn: https://www.linkedin.com/in/clairevo/ X: https://x.com/clairevo _Production and marketing by https://penname.co/._ _For inquiries about sponsoring the podcast, email jordan@penname.co._

Steve KaliskiguestClaire Vohost
Mar 24, 202641mWatch on YouTube ↗

At a glance

WHAT IT’S REALLY ABOUT

Stripe’s Slack-triggered AI “Minions” ship 1,300 reviewed PRs weekly

  1. Stripe uses Slack-activated AI “Minions” that provision isolated cloud dev environments, run an agent loop, and often deliver a ready-to-review pull request from a single prompt.
  2. The biggest benefit is reduced “activation energy” in large organizations: work can start from natural collaboration surfaces (Slack/Jira/Docs) rather than requiring an engineer to manually set up context in an editor.
  3. Minions ride on top of strong developer experience investments—hosted devboxes, internal docs, CI/testing, and standardized workflows—which increase agent success rates in a massive codebase.
  4. Stripe handles high volume (about 1,300 AI-written PRs/week) by relying on robust CI, test coverage, and safe deployment practices, shifting human effort from authoring to review and product judgment.
  5. A second demo shows agents as economic actors via a “machine payment protocol,” where Claude spends small amounts on third-party APIs (browser sessions, search, mail) and provides an itemized receipt plus a Stripe Climate offset.

IDEAS WORTH REMEMBERING

5 ideas

Lowering activation energy can matter more than faster execution.

Stripe starts work where ideas already live (Slack/Jira/Docs) and triggers agents with an emoji reaction, reducing the friction between a suggestion and a concrete code change.

Great DX is a prerequisite for reliable agents in large codebases.

Internal docs, “blessed paths,” and standardized workflows make it likelier an agent can find the right place to change code without blowing the context window or wandering.

Cloud environments unlock true multi-threaded agentic engineering.

Instead of running multiple worktrees locally, Stripe provisions many isolated devboxes in parallel, allowing multiple agents to work concurrently without laptop constraints.

One-shot prompting is feasible when the harness and tools do the heavy lifting.

The visible system prompt is intentionally minimal (“Implement this task completely…”) because the agent loop, tool access (MCP servers), and prebuilt workflows supply structure and context.

Review scalability depends more on CI confidence than on who wrote the code.

Stripe emphasizes strong test coverage, synthetics, and safe rollout/rollback (e.g., blue/green) so reviewers can trust changes whether authored by humans or agents.

WORDS WORTH SAVING

5 quotes

At Stripe, we're landing about thirteen hundred PRs that have no human assistance besides review per week.

— Steve Kaliski

I don't remember the last time I started work in the text editor.

— Steve Kaliski

When you're in larger organizations, there's so much friction that can come between a good idea and getting it into the world.

— Claire Vo

Whether the text has been written by Steve or the text has been written by Steve's robot, you still want that CI environment that's providing confidence.

— Steve Kaliski

You know you're doing something wrong if I have to load environmental variables to celebrate someone's birthday.

— Claire Vo

Minions concept and brandingActivation energy vs execution in large orgsCloud devboxes and parallel agent workstreamsAgent loop, system prompts, and one-shot PR creationGoose agent harness (forked/customized)CI/testing, blue-green deploys, and review at scaleMachine-to-machine payments and agent receiptsAPI-first, ephemeral businesses selling to agents

High quality AI-generated summary created from speaker-labeled transcript.

Get more out of YouTube videos.

High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.

Add to Chrome