Block has unveiled BuilderBot, an orchestration layer designed to manage multiple AI agents within its codebase, operating within Slack to allow users to describe needs for bug fixes or new features. According to Block, this setup enables real-time collaboration among team members, who can observe the bot as it researches, plans, and implements solutions. BuilderBot is built to understand the entire context of the codebase but strictly operates on source code and system configurations, ensuring that customer data, personally identifiable information (PII), and payment information remain untouched. Brad Axen, Head of AI Capabilities at Block, stated, "The best way to think about Builderbot is as the missing layer between AI coding tools and how engineering actually works at scale." On the Square side, Block's engineers delivered features that sellers had awaited for months in just days. BuilderBot is built on the open-source goose AI agent framework, which was the first contribution to the Agentic AI Foundation (AAIF) under the Linux Foundation. Over the past quarter, Block AI has also contributed to projects including TensorFlow and Hugging Face, and released several key datasets under permissive licenses.