Stripe Projects and Agent-based Development

8/7/2026

This post introduces agentic payments for marketplaces, detailing how AI agents can securely initiate and manage payments within marketplace contexts. It uses a smart fridge scenario to illustrate the technical challenges and solutions for enabling safe and automated financial transactions driven by AI agents, building upon the Stripe Projects framework and AI agent integration.

7/30/2026

Introduces Stripe's Knowledge AI Platform, a versatile AI agent platform designed for non-coding knowledge work. It connects employees to over 1,000 internal tools and skills, enabling secure, enterprise-scale productivity. The post highlights the platform's ability to handle diverse tasks from quick queries to complex, multi-day projects.

6/16/2026

This post focuses on the practical challenges of integrating AI agents with real-world systems, moving beyond the 'happy path' to address friction and error handling. It discusses how to get through these challenges when things go wrong, implying a focus on robust agent integration and error recovery mechanisms within the Stripe Projects framework.

5/26/2026

This post details the practical experience of building a GitHub traffic dashboard using Stripe Projects. It highlights the rapid materialization of infrastructure components like Auth0, Supabase, Vercel, Prisma, deployments, environments, and scheduled jobs. The author emphasizes the surprising shift from 'AI coding' to real infrastructure orchestration and the speed of development, while also noting emerging challenges in AI-assisted infrastructure such as operational visibility, ownership clarity, environment management, and understanding rapidly orchestrated systems.

5/14/2026

This post details learnings from a dozen steering experiments on how LLM-based agents behave in the wild, emphasizing the need for explicit instructions and structured prompting over implicit communication. It highlights that agents require clear, direct commands and structured inputs rather than relying on subtle cues or conversational nuances.

5/4/2026

This post details the practical learnings and insights gained from the process of building with agents and Stripe Projects, following up on previous posts that introduced the concepts and initial implementations. It focuses on the transition from initial development ('init') to deployment, highlighting the real-world challenges and solutions encountered.

4/29/2026

This post details a practical example of using Stripe Projects and AI agents to build an internal transcript editing application. It highlights the end-to-end development process, from initialization to deployment, integrating with external services like Vercel and OpenRouter. The post emphasizes the shift from pure 'AI coding' to real infrastructure orchestration within the Stripe Projects framework.

4/28/2026

This post details the development and rollout of rubyfmt, a Rust-based, zero-config, ultra-fast autoformatter, across Stripe's 25 million line Ruby codebase. It highlights a significant advancement in developer productivity and code standardization, demonstrating Stripe's commitment to maintaining code quality and consistency at scale, which complements the broader theme of agent-based development and infrastructure orchestration.

4/13/2026

This post details the application of AI agents, specifically CLI-based coding agents, to accelerate the development of Stripe-Salesforce integrations. It highlights the challenges of building these integrations due to the steep learning curve across both platforms and the tendency of AI agents to hallucinate on complex, evolving surfaces. The solution proposed involves using open-source grounding context files to flatten the learning curve and reduce hallucinations, thereby enabling faster development of working integrations.

3/26/2026

This post introduces the capability to provision a production-ready development stack from the terminal using Stripe Projects. It details how Stripe Projects can provision hosting, databases, auth, analytics, and AI tools by creating real resources in the user's provider accounts, securely syncing agent-ready credentials, supporting upgrades to paid tiers, and handling payments without leaving the terminal.

2/19/2026

This post details Stripe's 'Minions' system, which are one-shot, end-to-end coding agents that generate code from start to finish, with human review. It discusses the architecture and implementation of these agents, highlighting their role in automating a substantial portion of code generation and contributing to developer productivity.

2/9/2026

This post introduces 'Minions', Stripe's one-shot, end-to-end coding agents that handle the complete lifecycle of code generation, from initial writing to pull request submission. It highlights their scale (over a thousand PRs merged weekly) and their role in augmenting developer productivity by automating coding tasks.

1/27/2026

This post details the use of AI agents to generate Terraform code for Stripe configuration, highlighting the advantages of this approach over direct API calls for achieving transparent, consistent, and auditable infrastructure.

8/6/2025

This post details the integration of Hypermode's financial operations agents with Stripe, enabling users to build AI agents that can interact with Stripe APIs to perform financial operations using natural language. It highlights the use of Hypermode's platform to orchestrate these agents and leverage Stripe's capabilities for payments, refunds, and invoicing, effectively extending the Stripe Projects framework for AI-driven financial automation.

3/18/2025

This post details the technical implementation of Stripe's new AI Assistant in VS Code, focusing on retrieval mechanisms, evaluation strategies, and measurement techniques. It explains how the assistant leverages existing agent-based development frameworks to enhance developer productivity within the VS Code environment.

11/14/2024

This post details the integration of Stripe's agent toolkit with LLMs, enabling agentic workflows to perform payments. It covers how LLMs can be used to initiate financial transactions, manage metered billing, and access financial services, thereby expanding the capabilities of agent-based development within the Stripe Projects framework.