Case Studies

Three systems in production: what they do, and what they changed.

Diagram: many tangled flows leaving a column of systems, converging into ordered lines towards a second column.

Luxury cruises

Ponant / Talend

Project
Ponant's internal and external applications exchange their data through Talend flows, which we extend and keep in operational condition.
Impact
More reliable exchanges, fewer errors to fix by hand, and orders delivered within the expected times.
Stack
Talend Data Fabric 7.3/8.0OracleSQL ServerPostgreSQLREST/SOAPXML/JSONJava/J2EE
80+
virtual machines under supervision
0
downtime at release
24/7
metrics, logs and traces

Luxury cruises

Ponant / Kubernetes platform

Project
Design and operation of the Kubernetes platform (k3s) that hosts the line’s applications: databases driven by the cluster, continuous supervision, and automated delivery chains.
Impact
A release goes out in the middle of the day without anyone noticing, a database fails over on its own, and the alert reaches Teams before the user’s call does.
Stack
k3sCloudNativePGGrafana LGTM (Loki, Tempo, Mimir)n8nMicrosoft TeamsGitLab CILinux
Diagram: a scatter of dispersed records funnelled into one ordered register, with two chat bubbles above it.

Humanitarian NGO

LBAID, aid distribution tracking (ACA)

Project
The platform that tracks ACA’s aid distributions — families, campaigns, vouchers and receipts — with WhatsApp and Telegram chatbots for the field and an MCP server that opens the same operations to an AI assistant.
Impact
Family data is held in one place instead of scattered across spreadsheets, campaign tracking is automated, and distributors and donors see the same information at the same time.
Stack
PythonFastAPILangGraphMCP server (OAuth)PostgreSQLReactQR codesWhatsApp/Telegram/Web chatbots

Tell us what you want to automate.

One conversation is enough to know whether we can help, and how fast.

Talk to an expert+33 7 51 52 51 10