For the PM with a working prototype that keeps coming back rebuilt – or never coming back at all. The day engineering merges it instead.
You built the prototype. The customer said yes. The ticket goes in, engineering reads what you wrote, looks at what you made, and starts the build from zero. The third pass of the feature is the one that ships – the one nobody validated.
TL;DR
Most PMs who have shipped a validated prototype have watched engineering rebuild it from scratch. The prototype lived in a sandbox, the production code lives on the real frontend, and the translation between the two is engineering re-doing the work the PM already did.
Else runs the prototype on the real codebase from the start. When the prototype works, Else opens a frontend PR engineering reviews like any other. The artifact engineering reviews is the artifact the customer signed off on.
Else prototypes directly on your real codebase so engineering sees a pull request to merge, not a sketch to rebuild. The PM stops calculating whether the juice is worth the squeeze before ringing everyone’s bells, builds the change, and opens the PR. The prototype engineers will actually merge is the one that arrives in their inbox looking like a PR from the team.
Before Else
You built it yourself. A weekend in Lovable, or Figma Make, or Cursor on your second monitor. The customer you showed it to said yes. It works on your laptop.
When the ticket goes in, engineering reads what you wrote, looks at what you made, and starts the build from zero. If the ask is small – a copy update, a color, a three-pixel shift – it sits in the queue next to features nobody disputes matter more. A former PM at Escala described the meeting shape:
“And those are the most annoying things in our backlog when I was running Escala. It was a meeting with the marketing team, and the design team are all just waiting on these silly little tickets that eat up your engineering time that aren’t gonna move the needle.” – Matias Beeck, former Escala
A PM at HP named the structural problem under every version of this story:
“If we could get to a place where the prototype code is actually usable code, then that would be a huge time saver for engineering. And I think it would also mean that the final product is much more aligned with what we prototyped. Because there’s less rework and less reinterpretation of the design.” – Product Ops, HP
The case for building inside your real codebase instead of a sandbox is the case for a prototype engineering will merge instead of rebuild. When a prototype lives in a tool like Lovable or Bolt, what looks good in isolation becomes something engineers need to reinterpret. On your actual product, the prototype is the thing engineering merges.
That is the status quo – rework on both sides. Engineering rebuilds what the PM already built. The PM re-writes the prototype as a PRD so engineering can start over.
How Else does it
Three steps.
1. Install the GitHub app on your frontend repo. Else reads the repo, inherits the design system, and picks up the components and CSS already in production. The prototype looks like your product because it is running on your product. This matters more than it sounds – tools like Lovable or Bolt sit outside your codebase and require rework.
2. Build inside the actual product. Idea to working feature on top of the real app. Multi-page flows. UX reworks. Backend extensions handled alongside the frontend. Internal sharing links first, then targeted deployments to specific users or teams, then live user tests against real traffic when you are ready.
3. Open the PR engineering merges. When the prototype is ready, Else opens a frontend pull request on your repo. Engineering reviews, tests, and merges through the same pipeline they run on every other PR. The PR looks like any other PR, because it is.
No parallel codebase. No rewrite from screenshots. No ticket to rebuild what already works. This is what vibe-coding for PMs looks like when it stays inside your actual product – the prototype is the thing engineering merges.
Proof
Drew Muller, PM at Ferry International, has shipped two features into production through this path. Zero changes to the code Else produced before merge.
“With Else, we went from validated prototype to production PR in a fraction of the time it used to take. Engineering didn’t touch it until it was ready to merge.” – Drew Muller, Ferry International