Some apps get built fast, sometimes with AI coding tools such as Cursor or Claude Code, and find real users before they are ready for them. We take over existing apps and add the cloud setup, observability, DevOps, and security that get them into production and scaling reliably.
We take on each of these.
We review the codebase, the infrastructure, and how the app is deployed and run. We work out what works, what is dangerous, and what needs rewriting, and you get a prioritized plan with effort estimates. The plan puts keeping your business running ahead of a rewrite. If a rebuild is the faster path, we tell you, and the app you already have becomes the spec for the new one.
If production is failing, we fix that before anything else. Work that prevents the next outage goes ahead of work on the next feature.
We then work through the plan. That can mean a CI/CD pipeline with automated tests and security scans, and separate development, staging, and production environments. It can mean monitoring and alerting with tools like Sentry, Grafana, or OpenTelemetry, security fixes, and cost and abuse controls on AI features. Each change ships on its own instead of in one large rewrite.
We document the work and train your team. After that we can keep supporting the app or hand it off completely.
We wrote about the checks to run on an app like this, and when a rebuild is faster, in Securing a Vibe-Coded App, and about which security fixes fit which architecture in When Does a Vibe-Coded App Need RLS?.
“Graham moved us from playing with AI to production system in days. The capabilities are amazing but care in how he set us up was even better. The solution is persistent but respects privacy.”
$20M+ TVL under smart contracts.
Smart contracts we wrote and audited managed over $20 million USD in deposited funds across IDO launches, staking, and yield farming during the 2021 bull market, with the platform scaling to thousands of concurrent users on self-hosted infrastructure built specifically because cloud node operations were too expensive at that scale.
Case study
Live for years with zero exploits.
Smart contracts we wrote have managed real treasury funds across Ergo and Cardano for years with zero exploits. The Cardano expansion was funded by a Project Catalyst grant.
Case studyWe have rescued stalled projects where the previous team left and the codebase needed someone willing to read it carefully and figure out what was real. We have shipped production software since 2011.
Choose a slot in the booking calendar.
Say what it does, how it was built, and what is going wrong.
And what the work would look like. If we take it on, it starts with a fixed-scope triage, with the timeline agreed before we begin.
Yes. We take over existing apps, including vibe-coded ones. We look at the prior work without judging it and focus on what needs to happen next.
Only if a rewrite is the better path, and the triage shows which one it is. Sometimes the data model and business logic are sound and the app can be fixed in place. With smaller codebases or prototypes, a rebuild is sometimes faster.
Yes. Taking over an app after the previous team left, or when the original developers cannot be reached, is part of what we do.
Yes. We can set direction, review pull requests, and make the architecture calls while your team does the implementation.
Each fix ships on its own rather than in one large rewrite, so feature work can continue alongside it.
The triage is fixed-scope, with the timeline agreed before we start. Remediation is billed by time and materials, with weekly check-ins so you can change priorities as we go. You own the infrastructure code, Dockerfiles, and runbooks, and we deliver them as we build them.
