Flow Town: a map of the work behind software delivery
Flow Town is a free interactive map of Enablement Engineering: find where your software delivery is stuck, or choose your next DevOps specialisation.
Posts on CI/CD, test automation, developer platforms, and a few other things.
Some posts are in French: read them here
Flow Town is a free interactive map of Enablement Engineering: find where your software delivery is stuck, or choose your next DevOps specialisation.
I took the train to go from Paris 🇫🇷 to Stavanger in Norway 🇳🇴. That's 9 different trains, 2550km, about 33 hours on a train seat, and 3 days total.
A small open-source viewer that renders any note from your Obsidian vault as a local webpage, so bookmarks, todo lists and cheatsheets are one click away in your browser toolbar.
Merge queues automate rebasing and testing PRs at scale, with batching, speculative checks, and bisection tradeoffs explained through failure-rate math.
Fix slow CI with test selection: test impact analysis, only-changed flags and Meta's Predictive Test Selection skip the tests a change can't break.
A discussion on CI/CD strategies to keep main green: block bad changes with a merge gate, or land them fast and chase failures with automated bisection and rollback. Or blend both!
Slow CI and flaky tests feel like the price of quality. DORA research shows speed and stability move together. A series on test automation strategies.
A reproducible AI-assisted workflow for large-scale migrations across many repositories, with Renovate and Claude. 56 repositories, 5 hours, $80.
AI threatens millions of jobs while political debate ignores it. Four economic models for the transition: bans, UBI taxes, token taxes, job-creation incentives.
Tech giants and governments are converging on mass surveillance, from Chat Control to AI-powered facial scans. Why privacy is becoming a luxury.
Part 4/4: test the Koa.js, GraphQL, and MongoDB server with Mocha and Chai, using an in-memory MongoDB instance to keep tests fast and isolated.
Part 3/4: build a GraphQL API with Apollo Server and MongoDB, using GraphQL Code Generator to keep schema, types, and database models in sync.
Part 2/4: containerize the Node.js server and a MongoDB database with Docker and Docker Compose, then manage secrets with environment variables.
Part 1/4: bootstrap a Node.js server with Koa.js and TypeScript, then add ESLint, Husky git hooks, and Nodemon for a clean development workflow.
Learning Git - 101
Loading of a C++ class from a shared library (Modern C++)
Vivatech event review
Arch Linux - XFCE configuration