Make your AI app production-ready, spot technical risks before real users

• Auth: roles and permissions model, session handling, no open endpoints
• Database: tenant isolation, RLS and policies, no unprotected access
• APIs: validation, rate limiting, error handling on open interfaces
• Deployment: reproducible builds, separated environments, no hand deploys
• Monitoring: alerts and health checks so outages are not reported by users first
• Backups: regular backups and a tested recovery path
• Logging: traceable error trails without secrets or personal data in plain text
• CVEs: known vulnerabilities in the packages and dependencies you use
• Rollback: a defined way back to the last working version
• Secrets: no API keys or credentials in the frontend or the repository

1. 01

   Fit check

   A short, free conversation about stack, tool and current state. This clarifies whether a Baseline or ongoing support fits the phase before launch.

2. 02

   Scope and access

   We define the review scope and set up read-only access to the repository. For hosting, deployment and monitoring we look at the configuration together with the team.

3. 03

   Technical analysis

   I review the path from demo to production: backups and a tested recovery, monitoring and alerts, logging without secrets in plain text, secret handling, scaling under load and ongoing operation. Known CVEs in dependencies are included.

4. 04

   Report and recommendations

   Findings come prioritised with a risk rating and concrete recommendations, classified as fix now, fix before launch or plan for later.

5. 05

   Next step

   From the baseline it follows whether an ongoing plan like Oversight, Guard or Launch makes sense, so production readiness is held after launch instead of becoming an open question again.

Many projects start with a Baseline review. If the product is then developed further with AI, I can support it continuously.

• What does production-ready actually mean in concrete terms?

  To make an AI MVP production-ready means the app runs reliably not only in the demo but also under real users, load and failure cases. That includes access protection, a tested recovery, clean error behaviour, monitoring and a rollback path. We review exactly these points before real users and rank every finding by severity.

• Is checking once before launch enough?

  For the launch itself a Baseline is a sensible starting point. But as soon as the app is live, the state shifts with every feature and every new CVE. That is why we treat production readiness as an ongoing operations question: make it production-ready once, then keep it under ongoing technical oversight with Oversight, Guard or Launch.

• Do you do the fixes yourselves?

  Not within the plan. We review, prioritise and explain what to do before launch. Implementation runs separately through Wevelsiep Advisory or WZ-IT, or your own team. That keeps the review independent from the implementation.

• Do you need repo and infrastructure access?

  For the review, read access to the repository is usually enough, read-only by default. For infrastructure, deployment and monitoring we look at the configuration together with you. We do not need write access, because we do not commit the fixes ourselves.

• Which plan fits an upcoming launch?

  For the active launch phase, Launch at 3.900 € per month is the fit: ongoing oversight plus a pre-release check before each release and a direct channel with a short response time. Before that we recommend a Baseline at 790 € as an assessment. Stable products with little movement are well served by Oversight from 990 €.

• What does it cost?

  The entry point is fixed: Baseline 790 € as a one-off review. Ongoing oversight starts at 990 € per month (Oversight), then Guard at 1.950 €, Launch at 3.900 € and Scale on request. All prices net plus VAT. Ongoing packages start with a 3-month minimum term, then cancelable monthly, unless agreed otherwise.