Factory

Overview

Factory is an AI-native startup highlighted as an example of an AI-first operating model, where work is delegated to AI agents across functions rather than handled solely through traditional human workflows. In newsletter mentions, Factory appears both as a company building agentic tooling—such as Droid used through the Ghosty CLI—and as an organization shaping new AI-powered product management workflows and skills.

For AI Product Managers, Factory matters because it illustrates what happens when agents become part of day-to-day execution across engineering, product, design, and adjacent functions. Instead of treating AI as a lightweight assistant, Factory is presented as embedding high-autonomy agents into planning, specification, implementation, QA, and decision support. That makes it a useful reference point for PMs designing AI-first teams, operating rhythms, and product development systems.

Key Developments

• 2026-02-14: Factory was cited alongside Ramp and Linear as an AI-native startup that delegates tasks to AI agents across engineering, PM, design, and sales, with humans focusing more on context, systems, and feedback loops.
• 2026-02-15: Peter Yang previewed Eno's new AI-powered product management skill, built at Factory, which reportedly included product principles, positioning guidance, 11-star experience guidelines, PRD templates, review rubrics, and prioritization frameworks.
• 2026-02-16: Peter Yang demonstrated Factory's Droid agent via the Ghosty CLI in high-autonomy spec mode, using Opus 4.5 for planning and GPT-5.2 for execution to build and QA a React-based speed-reading web app. The workflow included spec generation in VS Code, Chrome DevTools for automated screenshots, and quality checks such as linting and type-checking.

Relevance to AI PMs

1. A model for AI-first team design Factory gives PMs a concrete example of how to reorganize work when agents can handle substantial execution. The practical takeaway is to redefine human roles around problem framing, constraints, prioritization, and review loops rather than around manual document creation or task-by-task coordination.

2. A blueprint for agentic product workflows
The Droid + Ghosty CLI example shows an end-to-end flow where an agent can gather inputs, draft an editable spec, implement features, and run QA checks. PMs can use this pattern to design workflows that connect planning, prototyping, validation, and iteration in a single agent-assisted loop.

3. A template for codifying PM judgment into reusable skills
Factory's work on AI-powered PM skills suggests that product principles, positioning frameworks, PRD templates, and review rubrics can be embedded into AI systems. For PMs, this is tactically useful as a way to scale quality, onboard teams faster, and make decision standards more consistent.

Related

• droid: Factory's AI agent referenced in a coding and product workflow demo; used for planning, implementation, and QA.
• ghosty-cli: The interface through which Droid was used in high-autonomy spec mode.
• opus-45: The planning model used in the demonstrated Factory workflow.
• gpt-52: The execution model used alongside Opus 4.5 in the Droid workflow.
• eno: Associated with the AI-powered product management skill built at Factory.
• ramp: Mentioned alongside Factory as another AI-native startup delegating work to agents.
• linear: Also cited with Factory as an example of an AI-native operating model.
• peter-yang: The newsletter source who highlighted Factory's workflows, tools, and product management relevance.