Blog
Notes from the Butterneck team
How product teams hand change requests to AI, and why a developer still decides what goes live. RSS feed.
Butterneck is a tool for AI change requests. A product person requests a change to a live web app and checks it on a live preview: the AI coding agent the developer already uses runs on the developer's computer, Butterneck does not commit, push or open a pull request, and a developer decides first. It is not a project-management or task-board tool.
Lovable alternative for teams with an existing app: an honest comparison with Butterneck
AI change requests as a Lovable alternative for the app you already have: PMs ask on a live preview, an AI coding agent edits, a developer decides what ships.
AI app builder or change the app you already have? How product teams should choose
AI change requests vs AI app builders: change the app you already have. PMs ask on a live preview, an AI coding agent edits, a developer decides what ships.
Agentic coding for product teams: what it is, and how non-developers can use it
AI change requests for agentic coding: PMs and non-developers ask on a live preview, an AI coding agent edits the project, a developer decides what ships.
Vibe coding for teams: how to let the whole product team ask for changes, safely
AI change requests for teams doing vibe coding: PMs and designers ask on a live preview, an AI coding agent edits, and a developer decides what ships.
Adaptl vs Butterneck: an honest comparison
How Adaptl and Butterneck compare: who opens the pull request, what data the preview shows, where it runs, Jira, and pricing. Written by the Butterneck team.
Can product managers change the app without engineering?
Short answer: PMs can make and check small app changes themselves with an AI agent, but engineering should still decide what ships. Here is a setup that works.
AI change requests in Jira: how to start one from a ticket
How to turn a Jira ticket into an AI-made change the requester can check on a live preview, with a developer deciding what ships.
How can non-engineers ship frontend changes as pull requests?
A 5-step workflow for designers and PMs: make the frontend change on a live preview, then a developer reviews it and opens the pull request.
AI tools that let non-developers change an existing web app (2026)
AI tools that let PMs and non-developers change an existing web app (2026): prices, where the agent runs, PR or live preview, and who decides what ships.
Live preview before the PR: how PMs can review a change before anyone opens a pull request
AI change requests with a live preview before anyone opens a pull request: what the PM checks, who decides, and how it differs from PR previews.
A 4-week pilot to let PMs request live changes: the proposal to send your engineering lead
A copyable one-page proposal for a 4-week pilot where PMs ask an AI coding agent for small changes on a live preview and a developer decides what ships.
How to write a change request for an AI coding agent: a template for PMs
A five-line AI change request template: how PMs and designers can prompt an AI coding agent to change an existing app, and what to leave to a developer.
Claude Code for product managers: use your team's Claude Code or Cursor without a terminal
AI change requests for PMs using Claude Code, no terminal or repo: ask on a live preview, your developer's Claude Code or Cursor edits, a developer decides.
Clogless alternative: an honest comparison with Butterneck
Clogless vs Butterneck for product teams: cloud sandbox and PR vs a live preview your developer reviews. Plus other Clogless alternatives, from public info.
The AI PR review bottleneck: why AI pull requests need an owner (2026 data)
The AI PR review bottleneck in numbers: LinearB (8.1M PRs), CircleCI, DX and a 196,212-PR CMU/Stanford study. Why AI PRs wait, and what an owner changes.
What non-engineers can ask AI to change: a policy template for product teams
A copyable policy for PMs, designers and support leads who ask an AI coding agent for changes: what they can ask for, what needs an engineer, and review tiers.
Should product managers ship code? A better middle ground
Should PMs ship code with AI? A Harvard study finds AI agents slow code review. A calmer option: PMs ask on a live preview, and a developer decides what ships.
Itera alternative: an honest comparison with Butterneck
How Itera and Butterneck compare for product teams: scope, where the agent runs, PRs vs previews, Jira and pricing. Plus other Itera alternatives.
Small changes stuck in the backlog? How PMs can change a live app without waiting on engineering
AI change requests for product managers: ask for small UI changes on a live preview without waiting for engineers. A developer still decides what ships.
Jira ticket to pull request with AI: check a live preview before anyone opens the PR
AI change requests from a Jira ticket: a label starts the work, the requester checks a live preview, and a developer decides whether to open the PR.
AI change requests for product teams: a practical guide
What AI change requests are, how product teams automate the request-to-preview handoff safely on a live copy, what guardrails to check, and a rollout plan.
How Butterneck runs an AI coding agent in a sandbox on your own machine
Run Cursor, Claude Code or Antigravity on a developer's computer for AI change requests: folder isolation, .env guardrails, per-turn rollback, preview links.
Ask freely, ship deliberately: an architecture for human-in-the-loop AI product changes
AI change requests with a human-in-the-loop architecture: the requester checks a live preview, Butterneck opens no PRs, and a developer decides what ships.
Human-in-the-loop AI coding: a developer decides what ships
AI change requests with human-in-the-loop AI coding: the person who asked checks a live preview, Butterneck never merges, and a developer decides what ships.
Share your local app with stakeholders without deploying it
AI change requests on a shared local app: PMs open a live link to the app on a developer's computer, ask in chat, and a developer decides what ships.