How a Refactor engagement actually runs

Five steps. Fixed prices. No surprise rewrites. Below is exactly what happens from the first call to the day you stop needing us.

1. 01

   Day 0 — 30 min, free

   TRIAGE CALL

   Share a link to your project (Lovable, Bolt, Cursor, GitHub, a screen recording — anything). A senior engineer reads it live and tells you, in plain English, what's broken, what's risky, and what's actually fine. If we can't help, we say so on the call.

   → A written summary of the call + a fixed quote, in your inbox the same day.
2. 02

   Week 1 — fixed price

   AUDIT REPORT

   We spend 3–5 days reading every file. Security, data model, auth, performance, maintainability. Every finding gets a severity (P0–P3), a location in your repo, an estimated fix time, and a recommended fix. Nothing vague.

   → 18–25 page PDF + GitHub issues filed in your repo + a 60-min walkthrough call.
3. 03

   Weeks 2–4 — fixed price

   REFACTOR SPRINT

   We work in your repo, in branches you control. PRs go up daily. Each PR has a description a non-engineer can read. Nothing gets rewritten that doesn't need to be. You keep shipping the whole time.

   → Merged PRs you understand, a green test suite, no surprise rewrites.
4. 04

   Inside the sprint

   HARDENING

   Row-level security on every table. Secrets out of the client. Rate limiting. Error reporting wired in. CI/CD that fails loudly. Backups that actually restore. The boring stuff that decides whether your launch survives traffic.

   → A system that doesn't leak data, money, or your weekend.
5. 05

   Final week

   HAND-OFF

   Runbook for the 3 things that break most. README that takes a new hire from clone to running app in under 5 minutes. ADRs explaining anything weird. A recorded walkthrough so you don't depend on us to keep going.

   → You own everything. No lock-in, no retainer, no surprises.