How we work
Four steps from your first message to code in your repository. Nothing here is a discovery workshop.
Send a paragraph, a repository, a Figma file, or a half-finished spec. If you want an NDA signed before you send anything, say so in the first line and we will send ours within the hour.
Thirty minutes with the person who would write the code. We will have read whatever you sent. Expect questions about failure modes and data volumes rather than about your budget range.
A written scope with sequencing, the risks we can see, an estimate with its assumptions listed, and a section on what we would not build yet. It is yours whether or not you hire us — several people have taken it to another firm, which is a fair outcome.
A defined, useful slice of the work at a fixed price. You get to judge us on shipped output before committing to anything longer, and we get to find out what the codebase is really like.
Three things we commit to.
Async by default, in your tools. Slack Connect, your Linear, your GitHub. The overlap window is 13:00–17:00 IST, which is 08:30–12:30 in London and 17:30–21:30 in Sydney, each moving by an hour when their clocks change. We will move it if your mornings are what matter.
The most common complaint about outsourced engineering is what arrives at handover. So it is scoped work, not goodwill: an architecture note per subsystem, a runbook per service, an ADR for every decision with a stated cost, and a recorded walkthrough.
Written as we go. A doc produced in the last week of a project is a doc nobody trusts.
Your repositories and your cloud accounts from day one — we work inside them, never in ours. Handover is a scheduled two weeks: your engineer deploys, we watch. Then credentials are rotated and we are out of the accounts.