---
title: "SWFT · Software factories and agent swarms"
canonical: "https://swft.io/"
updated: "2026-09-17"
---

# SWFT

How organizations put agents to work at scale.

SWFT tracks software factories and agent swarms across industries: how organizations put agents to work, their reported scale, and the evidence behind the results.

Explore [company cases](/companies), [software factories across industries](/industries), and [agent scale](/scale).

## Source-led field analysis

- [Three visions of the AI software factory](/ideas/three-visions-of-the-ai-software-factory) — AI-drafted by SWFT Editorial from cited public sources. A practical guide to combining bounded production lines, Huntley's embedded Ralph loop, and Yegge's multi-agent work graph.

## Ben Guo's personal essay

- [The rise of software factories](/the-rise-of-software-factories) — Authenticated author-owned work by Ben Guo. A personal manifesto about building more software by improving the system around coding agents.

## How the work gets done

AI-drafted, source-led accounts of what changed, how the work moves, and where people still decide.

- [Jane Street AIDE: making agent work easier to verify](/companies/jane-street-aide) — A trading firm's approach to coding agents puts the development environment and the quality of feedback at the center of the story.
- [Stripe Minions: how coding tasks become reviewed changes](/companies/stripe-minions) — An employee delegates once. Minions prepares a development computer, changes code, runs checks, and returns the result for two stages of human review.
- [Shopify River and Aquifer: a shared platform for agents across the company](/companies/shopify-river-aquifer) — Employees work with River in shared Slack threads. Shopify's Aquifer design keeps the work history separate from temporary computers so sessions can survive interruptions.

## Compare 24 company cases

[Open the company comparison](/companies) for the main job, reported use, human checkpoint, public access, and evidence date in one non-ranked view.

[Across industries](/industries) follows the work beyond software vendors. [Agent scale](/scale) separates simultaneous workers, accumulated runs, output, adoption and workload, with sources and limits for each observation.

## The software factory field guide

- [What is a software factory?](/what-is-a-software-factory) — A plain-English guide to the people, agents, checks, and learning loop behind an AI software factory.
- [How an AI software factory works](/software-factory-architecture) — How requests reach agents, how results are checked and released, and how the team learns from the work. A practical map of the system around the model.
- [How to build a software factory](/how-to-build-a-software-factory) — Choose a recurring job, compare it with today's process, prepare the agent's tools and tests, and measure what it takes to deliver an accepted result.
- [Software factory metrics that matter](/software-factory-metrics) — Five numbers for judging speed, rework, quality, human attention, and the real cost of AI-assisted delivery.

[Browse every guide and reference](/guides).

## Publication

SWFT is free to read and source-linked. Ben Guo's authenticated personal essay is labeled separately from the AI-drafted guides and cases published by SWFT Editorial. [About the publication](/about), read the [editorial method](/methods), or follow the [Atom feed](/feed.xml).

Homepage selection date: 2026-09-08.