Find out what is broken before it breaks you. I go through your system and hand you a clear report.
Arrives by email in a minute. No card needed.
Experience from
Free Start here
A quick scan for each area. Your own AI runs it on your project and shows you the first real problems it finds.
€0
One or all four. No card needed.
Do it yourself
Includes a code scanner your own AI runs on your project, checks for what AI-built apps most often get wrong, and writes a report with a fix prompt for every finding.
€149€196Save €47
All four skills €37.25 each
or €49 per skill
Done for you
I go through your system myself, with the scanner as a starting point, and hand you a clear, prioritised report. Pick one area or all four.
€899€1196Save €297
All four audits €224.75 each
or €299 per audit
Bespoke package
Once the audit is done, I can implement the fixes myself. Backend, infrastructure, security hardening, architecture improvements. Also for systems that are broken, unclear or risky: urgent audits, production debugging and pre-launch reviews.
from€1,499
Book a callFour areas, each with a report from one audit of a working app that was built with AI. Names and identifiers are changed. Every finding has evidence, a severity and a fix prompt.
Vulnerabilities, authentication flaws, secrets exposure, data access risks.
0 critical, 2 high, 3 medium, 2 low
Read the exampleCloud setup, deployment reliability, cost efficiency, incident readiness.
4 critical, 3 high, 4 medium, 1 low
Read the exampleSystem design, tech debt, coupling, bottlenecks, long-term maintainability.
0 critical, 2 high, 1 medium, 3 low
Read the exampleLoad handling, database performance, caching, growth readiness.
1 critical, 1 high, 2 medium, 4 low
Read the example10+ years building large-scale backend systems in public sector, fintech and research. I led architecture for borger.dk and built features at Public.com. My MSc from DTU's Honours Programme (top 10%) was published in PLOS ONE. I know what breaks under pressure because I have had to fix it.
Watch the video
This loads content from LinkedIn, which may set its own cookies.
A skill file you install in Claude Code or Claude.ai, or paste into another AI assistant such as ChatGPT or Cursor. You get it by email. The free version runs a quick scan of your project and explains the first real problems it finds. The full version adds a complete scanner, a manual checklist, stack-specific checks and a fixed severity scale. It writes a report with a fix prompt for every finding that you can hand straight to your AI assistant, and can re-check your project after you fix things.
The scanner only reads files. It runs on your own machine and sends nothing anywhere. The one exception is an optional online dependency check, which asks you first and sends only your package names, not your code, to the npm registry. The skill writes its report into your project folder and changes nothing else unless you ask it to fix findings, and then it does so one at a time with your agreement. Your AI assistant reads your files to do the audit, so what it does with them follows that tool's own privacy terms.
Founders, small teams and engineers who build products with AI and want an independent check on security, infrastructure, architecture or scalability. It is not a certified penetration test or a compliance sign-off such as ISO 27001 or SOC 2. If you are not sure it fits, book a call and ask.
Pick the skill if you want to run the audit yourself, on any codebase, any time. Pick the audit if you want me to go through your system and give you my judgement. You get a clear, prioritised written report with findings and next steps within 2-3 days.
Yes. The bespoke package covers backend, infrastructure, security hardening and architecture improvements, starting from €1,499. Book a call and we will scope it.
Start with one free audit skill. It takes a few minutes and shows you what to look for.