
Understand
We examine the product, its access, useful data and the relevant journey. You receive a shared assessment and a list of the unknowns that still prevent a quote or decision.
You show us what exists or what is missing. We then define the next decision, what will be delivered and how the result will be checked.

We examine the product, its access, useful data and the relevant journey. You receive a shared assessment and a list of the unknowns that still prevent a quote or decision.

We agree the next step, what is included, what is excluded and the criteria used to accept the result. You know what will be delivered now, what will wait and what still needs to be provided.

The functional prototype, code, integrations and deployment progress through versions you can review. Each version shows a usable part and lets the workflow be corrected before the scope grows.

The selected tests, access, configuration and documentation are brought together with the delivery. You have the material needed to operate, maintain or hand the next step to another provider.
For Creation, we start with the service to provide and a first version to test. For Recovery, we start with the files, code, access and deployment that are actually available.
An existing product on another stack can still be reviewed when its code, accounts and deployment environments are accessible.
Models and gateways are selected for data sensitivity and the required level of control: managed hosting, local execution or multiple providers.
Cloudflare Workers AI
Managed inference endpoints operated by Cloudflare, at the network edge.
Visit siteComfyUI
Node-based visual generation pipeline, for prototyping or producing assets without writing code.
Visit siteGemini
Google multimodal family, handy for reviews that mix text, screenshots and code snippets.
Visit siteHiggsfield
Reference image and short video generation, when a visual is missing and an honest stand-in is needed.
Visit siteOpenCode
Multi-agent terminal to code, refactor and test on a local machine or an ephemeral runner.
Visit siteBased on the risks observed, these tools protect access, harden exposed surfaces, restore a traceable deployment and measure availability.
Bitwarden
Password and secrets vault, to bring scattered access back together and renew whatever has circulated.
Visit siteDocker
Containerised environments, to rebuild an existing system whose original machine is no longer documented.
Visit siteGitHub
Code repository, reviews and continuous integration, when recovery relies on the history and deployment of an existing project.
Visit siteGitLab
Code repository, reviews and continuous integration, when recovery has to restore a traceable history and deployment.
Visit siteGnuPG
Signing and encryption of files and messages, notably to hand over secrets during an access recovery.
Visit siteLinux
Servers and runtime environments where patches, updates and access restrictions are applied.
Visit siteProton
Encrypted services suite (mail, calendar, drive, VPN) to take back control over sensitive data and access.
Visit siteProton Pass
End-to-end encrypted password manager, part of the Proton ecosystem, to bring scattered access back together.
Visit siteSyswarden
Server and service monitoring with a public status page, to measure availability after a recovery.
Visit siteUptime Kuma
Self-hosted availability monitoring, to measure outages and check the effect of fixes.
Visit siteYubico
Hardware security keys (YubiKey) to authenticate sensitive access without passwords or time-limited codes.
Visit siteDescribe the uncertainty, the product concerned and what is currently missing to decide without committing budget in the wrong direction.
Present the decision