architecture - development - operations



With 20 years in the opensource field, we accumulated a large experience in systems architecture design.
We work under the devops mantra: Culture, Automation, Measurement, Sharing.

Architecture & design

Web stack architecture, capacity planning, scale-out. Infrastructure provisionning, whether on-premises or {C,I,S,P}aaS/cloud.

Deploys & operations

Internet stack deployment. Operation and supervision 24/7/365 with post-mortems analysis and monthly operations reports. You have access to dashboards and everything is transparent. Hot-potatoes, skeletons in the closet and NotMyProblem attitude is not in our culture.

Performance optimisation & high-availability

Performance optimisation, leveraging caches, clusters and configuration "tuning", fed back by measurements. Improving uptime through highly-available setups.

Automation & measurement

Infrastucture automation (bootstrapping/deploy/management). CI/CD/CD pipelines setup. Infrastructure supervision, metrology, logging and APM (application performance management).

Tools we love (and run in production)

We work mostly with Ansible, Rancher, Docker, EF/LK (Elasticsearch, Filebeat/Logstash, Kibana), HAProxy, Varnish, Nginx, various MySQL versions (Percona Server/XtraDB, MySQL, MariaDB & Galera), Memcached/Redis, Corosync/Pacemaker, Redis/Sentinel, Cyrus-IMAP, Docker. And yes, all of these in production. We love SaaS too, and use Git{hub,lab}, DataDog, PagerDuty, StatusCake, AWS, GCP, Vultr, DigitalOcean, Uptimerobot.

What we don't do

We do not work on non-OSS stuff, although we can help you transitioning from proprietary software to open source. We do not work in "hey put this in production, thanks" mode. We work with upstream teams to create the best workflow, so we can ship qualitiy code fast. We don't build "web sites" (we did this one, you get the idea...) although we have friend companies we can recommend that excel in this area.


Shipping to production should be easy and a non-event. We design, build infrastructure and provide tooling so developpers can deploy to production seamlessly, without worries and can get a constant feedback on the application behavior.


Private CaaS architecture

Docker-based private CaaS

Elasticsearch capacity planning

Capacity planning for large indexes

Infrastructure Automation

Automated provisionning & deploys

High traffic web site operations

Deployment, management & operations

Large scale email infrastructure

SMTP/IMAP stack large mailstore

Dashboard development

Specific Technical Develoment

Architecture, deployment & operations


Large Network Design and Deployment

Project CHARM

Proxy Cluster Deployment



All things Ansible, networking, webstacks. Linux addict since the SLS days, never looked back. Frequently found in his sleeping bag in the wild.


The Boss. Handles financial, accounting, contracts and everything that geeks are allergic to. As a PagerDuty backup, she pokes the boys with a very pointy stick if her phone rings.


Sysadmin & network freak. RFC1149 expert. KISS philosophy aficionado. Plays with old electronic boards, listens to weird music at the office and enjoys nice smelly cheeses.


Sysadmin & Devopsy stuff. When he's not playing bass or making ice-cream, he looks after CI/CD pipelines, does weird stuff with kubectl while playing double-pedal-air-drums in the office.