ia
Deep Dive: Generating AWS diagrams with an AI agent and Draw.io MCP
How I use an AI agent with the Draw.io MCP to automatically generate professional AWS architecture diagrams, directly in Draw.io.
jls42.org
Personal blog about AI, infrastructure and tech projects
Recent Articles
All posts 
blog
New logo and migration to Astro: jls42.org gets a facelift
Discover the new logo incorporating AI symbolism and the complete migration of the blog from Hugo to Astro, with automatic translation into 15 languages.
blog / ia
Babel Fish AI: New Features, Context Menu, Auto-Copy, Code Quality Control, etc.
With Babel Fish AI, I previously introduced a Chrome extension able to convert voice to text via OpenAI's Whisper API, also offering real-time translation. Today I'm back with a batch of new features and many improvements: from the context menu to automatic copying of the transcribed text, including automatic rephrasing (now activatable from a button in the banner), this post takes stock of the changes since the very first version.
blog / ia
Launch of Babel Fish AI: Chrome Extension for Voice Transcription and Translation
Babel Fish AI, a personal project, is an innovative Chrome extension that converts voice to text with exceptional accuracy, while offering an optional automatic translation feature. Designed to be reliable and ad-free, it provides high-quality voice transcription via OpenAI's Whisper API. I created this extension partly to address a personal need: to simplify communication with AIs by dictating my requests.
infrastructure / blog
Update: LibreChat version v0.7.6 with automated deployment on AWS EC2
I'm happy to announce that my automated deployment project for LibreChat on AWS EC2 has been updated to fix issues related to recent changes in how LibreChat is installed, while now supporting version v0.7.6.
blog / ia
Integrate the OpenAI o1-preview model into a Python script with LaTeX expression handling
Discover how to integrate the new OpenAI o1-preview model into a Python script to enhance your AI projects. This script allows you to interact with the OpenAI API using the o1-preview model, with the ability to include web content in prompts via a web scraping feature. It also correctly handles LaTeX math expressions in the model's responses, converting them into readable Unicode text in the terminal.
Latest AI News
All news
Claude subscription policy revamped, Claude Code v2.1.92, Veo 3.1 Lite and Lyria 3
April 3-4, 2026: Anthropic removes third-party tool access from Claude Pro and Max subscriptions with compensatory measures, publishes Claude Code v2.1.92 with a Bedrock wizard and per-model cost details. Google launches Veo 3.1 Lite in the API, Lyria 3 in the Gemini app, and expands video generation in Workspace.
Microsoft 365 integrated into all Claude plans, Copilot signs its commits and Gemini CLI 0.36
On April 3, 2026, Anthropic opens Microsoft 365 connectors to all Claude plans, GitHub strengthens the Copilot cloud agent security with signed commits and organization-level firewall, and Gemini CLI moves to version 0.36.0 with native sandboxing.
Gemma 4 open source, Qwen3.6-Plus leads agentic coding, Anthropic explores functional emotions in LLMs
April 2, 2026: Google releases Gemma 4 under Apache 2.0 with native vision and audio, Qwen3.6-Plus tops Terminal-Bench with a 1M-token context window, and Anthropic publishes foundational research on functional emotions in language models.
Claude Code NO_FLICKER and v2.1.89, Copilot /fleet, Perplexity Computer in Slack
April 1, 2026: Claude Code v2.1.89 with NO_FLICKER and Computer Use, Copilot CLI /fleet for parallel agents, Perplexity Computer integrated in Slack, Codex CLI 0.118.0, GLM-5V-Turbo from Z.ai, GrandCode beats Codeforces, Runway Builders, and two April Fools' jokes.
OpenAI raises $122 billion, Qwen3.5-Omni and 15 hidden features of Claude Code
March 31, 2026: OpenAI closes a record financing round at a $852 billion valuation, Qwen launches its native omnimodal model, and Boris Cherny reveals little-known features of Claude Code.
Luma Uni-1 thinks and generates pixels, Perplexity on 1B Samsung devices, Claude Code v2.1.86
Week of Mar 26–28, 2026: Luma launches Uni-1 (unified visual reasoning), Perplexity powers Samsung Browsing Assist (1B+ devices), Claude Code v2.1.86 brings ~15 major fixes, GitHub Copilot CLI adds agent-driven unit test generation.
Projects
View project
Babel Fish AI: Voice Transcription and Translation Extension
Open-source browser extension for voice transcription and translation. Built with Roo Code, maintained with Claude Code and Gemini CLI. Chrome and Firefox.
Multilingual Automation: My AI-Powered Markdown Translator Script
Open-source Python script to translate Markdown files via the OpenAI, Mistral AI, Claude and Gemini APIs. A project born from my collaboration with AI.
Automated Deployment of LibreChat on AWS EC2: An Innovative DevOps Solution
Fully automated deployment of LibreChat on AWS EC2 with Terraform, GitLab CI/CD and AWS SSM. Operational instance in under 6 minutes with optimized costs.
StoryPixAI: Generate Children's Illustrated Stories with AI and AWS
Interactive web application that generates personalized children's stories with AI illustrations. Uses GPT-4, DALL-E 3, AWS Lambda and Terraform.