Articles

I publish articles on several media.

Online Technical Articles

I share and cross-post technical articles across three main platforms: the Worldline Engineering Blog, Medium, and DEV.to. Most recent works are hosted on DEV.to and organized into various thematic sections.

Vibe Coding one Step at a Time

This series documents the shift from traditional "bricklayer" coding to intent-driven development, exploring how to build tools and applications by focusing on outcomes while delegating implementation to AI.

• Vibe Coding One Page at a Time: A practical start to the series where a Python script is "vibed" into existence to solve a daily frustration: turning online magazine pages into high-quality, searchable PDFs using OCR and image optimization.
• Vibe Coding One Pixel at a Time: Moving from scripts to UI, this installment explores building a functional, vector-based SVG editor in the browser. It highlights "prompt anchoring" and using documentation as long-term memory for AI pair programming.
• Vibe Coding One Slice at a Time: The "Boss Level" of the series, demonstrating how to build a complex Modular Monolith. It introduces Governance-Driven Development (GDD), where architecture records and rules act as the "constitution" for AI development.

Beyond the ComfyUI Canvas

This series focuses on the programmatic side of ComfyUI, moving past the visual interface to explore automation and API integration.

• Unlocking ComfyUI's Power: A Guide to the HTTP API: An introduction to interacting with ComfyUI through its REST API, demonstrating how to trigger workflows and manage resources using Jupyter notebooks.
• WebSockets & ComfyUI: Building Interactive AI Applications: A technical deep dive into using WebSockets for real-time communication with ComfyUI, enabling live updates and interactive image generation feedback.
• Automating Image Generation with n8n and ComfyUI: A guide on bridging ComfyUI with the n8n automation platform to create powerful, low-code AI pipelines and generative workflows.
• Beyond the API: Integrating ComfyUI and Flowise via MCP: A look at the next frontier of AI orchestration by connecting ComfyUI to Flowise using the Model Context Protocol (MCP). This setup allows LLM agents to "see" and use your ComfyUI workflows as native tools within a chat-based interface.

Image Generation

• Enhancing QR Codes in the Age of GenAI: This article showcases practical examples of how tools like ControlNets enable the creation of QR Codes that are both functional and visually appealing through innovative techniques.
• 🎨JSON Style Guides for Controlled Image Generation with GPT-4o and GPT-Image-1: Learn how JSON style guides provide a consistent, modular, and precise approach to controlling image generation processes with GPT-4/GPT-Image and Gemini/Imagen.
• The Yoga of Image Generation: A three-part series exploring how Stable Diffusion and ComfyUI can be used to produce accurate, diverse, and inclusive images through mindful techniques.

Analyzing LLMs

• How Open is Generative AI?: An in-depth, two-part examination of the openness of various language models, exploring their different components and aspects.
• A comparative analysis of Chatbots/LLMs Performance across varied Prompts: This article presents an unconventional, hands-on benchmark comparing the performance of multiple LLMs across different prompt types.

Chat with Your Content

• Discover the new OpenAI Embeddings APIs: A comprehensive guide to understanding their features and integrating them using Node.js.
• RAG Tutorial: Exploring AnythingLLM and Vector Admin: Step-by-step instructions to deploy local models for text embedding and generation, including setup of a vectorstore for seamless integration.
• Reconquer your documents with Ragna: Ragna differentiates itself from generic frameworks with minimal abstractions, enabling faster prototyping via CLI tools while supporting scalable architecture.
• From Local AI to Enterprise-grade deployment with BionicGPT: Built in Rust, BionicGPT offers safety and high performance, featuring a robust architecture suited for industrial deployment.
• Enhance your Website with AI: Embed a GPT-ChatBot with Flowise: A detailed tutorial guiding you through adding a chat interface to your website by deploying Flowise on Hugging Face to create a functional ChatBot.
• In bed with GPT and Node.js: An extensive walkthrough on combining GPT with various vector stores and JavaScript libraries to develop more interactive and efficient applications.

Bringing AI Home

This series delves into open-source solutions for private, self-hosted AI tools, focusing on Large Language Models and AI assistants accessible from desktop environments.

• Introducing Jan: An exploration of Jan, an open-source alternative to LM Studio from Southeast Asia, offering a full open-source LLM experience.
• Introducing LocalAI: LocalAI simplifies running models inside Docker containers, making APIs readily available for local deployment.
• Introducing LM Studio: An overview of LM Studio, a convenient desktop tool for installing and testing open-source large language models.
• Introducing GPT4All: GPT4All is designed for those seeking a privacy-preserving, locally operable chatbot without demanding high-end hardware.

My Journey to AI

• No-code AI: OpenAI MyGPTs, LlamaIndex RAGs, or LangChain OpenGPTs?: An exploration on creating customized GPTs using OpenAI, along with what the RAGs and OpenGPTs projects from LlamaIndex and LangChain enable you to achieve with minimal coding.
• Open Source AI: Developing a Hugging Face ChatBot from scratch: A concise tutorial guiding you through building a chatbot with entirely open-source components, including the language model.
• Create a ChatBot with VertexAI and LibreChat: Learn to seamlessly combine Google's VertexAI with the open-source LibreChat platform to develop a advanced, ready-to-use chatbot.
• A Guide to Node.js Notebooks: An introduction to Jupyter Notebook’s capabilities for in-browser code execution and visualization, supporting multiple programming languages including JavaScript shells.

Web3

Web3 Unveiled: How Decentralization is shaping the future of the Internet: This article introduces Web3 as the anticipated next phase of the Internet, examining its core concepts, implications, and the ways it could transform our digital interactions.

JavaScript

• Object-Oriented JavaScript: More Than Just Classes: This article explores JavaScript's native object orientation rooted in prototypes, beyond the class syntax introduced in ECMAScript 6.
• The JavaScript Odyssey: A narrative chronicling the journey, innovation, and evolution of JavaScript through decades of development.
• The Ascent of Node.js: How a runtime changed the Web: A retrospective look at how Node.js emerged from its beginnings to fundamentally reshape web development, particularly on the server side.

Introducing SVG

• SVG - Open standard for vector graphics: An introduction to SVG as the open standard for creating static vector graphics suitable for web use.
• SVG - Dynamic vector graphics: This piece uncovers the lesser-known dynamic capabilities of SVG, enabling interactive and animated vector graphics.

Open Source

• An Open Source History: A historical overview tracing the development of Open Source from its origins to widespread adoption in the mass market.
• Open Source is Not Only About Code: This article highlights the various layers and dimensions of Free and Open Source Software beyond just the code itself.

Technical Publications

• Enhance Your QR Codes in the Era of Generative AI (French) published in Programmez! #270 (07/2025).
• A few years ago, I regularly published articles in the French magazines GNU Linux Magazine France and Linux Pratique. Some of these articles have been digitized and are still accessible on this platform.

Articles on Yoga

• I share articles related to Yoga in French on my website YogĀrkana.
• And occasionally in the French publication Le Journal du Yoga.