Charly Wargnier

31K posts

Charly Wargnier banner
Charly Wargnier

Charly Wargnier

@DataChaz

Hermes + 🦞 @Openclaw tinkerer • Ex @Streamlit @Snowflake Maestro • Tweet about AI agents, LLMs and Automation • My ❤️ is open source • DM for collabs 📩

London 🇬🇧 ⇆ 🇫🇷 Pyrenees Katılım Ocak 2009
48.5K Takip Edilen170.7K Takipçiler
Sabitlenmiş Tweet
Charly Wargnier
Charly Wargnier@DataChaz·
🚨 STOP WASTING MONEY ON AI API TOKENS. I recently discovered 9Router, and it completely changes the game. A 100% open-source router for Claude Code, Cursor, and Codex. Just set it up, and it quietly manages everything in the background. → Built on a brilliant 3-tier fallback system → Shifts to a budget-friendly model when premium quotas dry up → Defaults to a free model if your budget hits zero Plus, it slashes input tokens by 40% and balances multiple accounts automatically. ... it even translates requests across OpenAI, Gemini, and Claude flawlessly 🤯 Instantly compatible with 12+ coding tools. Get started here: `npm install -g 9router` 100% free and open-source. repo link in 🧵↓
Charly Wargnier tweet media
English
12
19
120
10.7K
Charly Wargnier
Charly Wargnier@DataChaz·
🚨 STOP WASTING MONEY ON AI API TOKENS. I recently discovered 9Router, and it completely changes the game. A 100% open-source router for Claude Code, Cursor, and Codex. Just set it up, and it quietly manages everything in the background. → Built on a brilliant 3-tier fallback system → Shifts to a budget-friendly model when premium quotas dry up → Defaults to a free model if your budget hits zero Plus, it slashes input tokens by 40% and balances multiple accounts automatically. ... it even translates requests across OpenAI, Gemini, and Claude flawlessly 🤯 Instantly compatible with 12+ coding tools. Get started here: `npm install -g 9router` 100% free and open-source. repo link in 🧵↓
Charly Wargnier tweet media
English
12
19
120
10.7K
Invincible
Invincible@InvincibleEdge·
@DataChaz @raycast finally. been watching mac friends flex this for years time to see if it lives up to the hype
English
1
0
1
4
Charly Wargnier
Charly Wargnier@DataChaz·
finally, someone using AI in a meaningful way
English
8
3
41
7.2K
Charly Wargnier
Charly Wargnier@DataChaz·
What a fascinating read! Miles’s productivity skyrocketed after making Claude his full-time assistant. He built a centralized personal OS for his tasks, finances, calendar, journal.. EVERYTHING It took weeks to perfect, but he’s gifting us the exact blueprint 👀↓
Charly Wargnier tweet media
Miles Deutscher@milesdeutscher

x.com/i/article/2056…

English
14
52
293
51.5K
Charly Wargnier
Charly Wargnier@DataChaz·
BOOM💥 Qwen3.7-Max drops on @AIMLapi and the stats are unreal. > 92.4 GPQA Diamond > Apex: 44.5 > 35+ hours of autonomous execution. Let that sink in. Go comment "Qwen" on their post, hit up their Discord, and claim your free access 👀
AI/ML API@aimlapi

Qwen3.7-Max on AI/ML API - built for the agent era GPQA Diamond (92.4), HMMT (97.1), Apex (44.5) Sustains 35+ hours of autonomous execution Works with Claude Code, Qwen Code & more Comment Qwen to get Free promo code

English
3
13
48
4.1K
Charly Wargnier
Charly Wargnier@DataChaz·
Full prompt below 👇 Create a complete single-file Super Mario Bros arcade game in pure HTML/CSS/JS, output ONLY raw HTML starting with and ending with , no explanations. Canvas is 800x500, centered on black page, body margin 0. Draw ALL graphics with Canvas 2D API using colorful pixel-art rectangles — no images. Background is sky blue (#5C94FC) with white cloud rectangles and green hill rectangles at the bottom. Ground is a row of brown brick blocks (#C84C0C) at y:440 h:60 spanning full width, with question-mark blocks (#FAB000 with ? in white) and brick blocks (#C84C0C) floating at y:280. Mario is 24x32px drawn with rectangles: red hat, skin face, brown mustache, red shirt, blue overalls, brown shoes — starts at x:60 y:408, moves left/right with arrow keys at speed 3, Space or Up jumps (vy=-13 only when onGround=true), gravity vy+=0.5 per frame, lands on any block when feet overlap block top and vy>=0. Goombas are 28x24px brown mushroom enemies that walk left at speed 1, reverse direction at screen edges, drawn as brown oval body with darker brown feet and angry white eyes — Mario kills Goomba by jumping on top (Mario vy>0, feet touch Goomba top), Goomba kills Mario on side touch. Coins are yellow 12x16px rectangles floating above ground, Mario collects by touching, score +100 each. Question blocks when hit from below by Mario (Mario head touches block bottom while vy<0): play bump animation, release a coin or mushroom, turn gray. Mushroom is red 20x16px rectangle that slides right, Mario grows bigger (48px tall) when collected. HUD at top: black bar, white Courier New text showing MARIO, SCORE (6 digits), COINS count with coin icon, WORLD 1-1, TIME countdown from 400. Game has TITLE screen with "SUPER MARIO BROS" in white, flashing "PRESS ENTER", and GAME OVER screen; Mario has 3 lives shown as mini Mario icons. Level scrolls horizontally — camera follows Mario, world is 3200px wide with varied block layouts, pipes (green rectangles 40x60px), gaps in ground that kill Mario if he falls in.
English
1
2
13
1.5K
Charly Wargnier
Charly Wargnier@DataChaz·
MIND BLOWN that an open-source model running on a MacBook can go toe-to-toe with the cloud 🤯 Spent yesterday in @atomic_chat_hq testing Gemini 3.5 Flash (Cloud) against Gemma 4 31B (Local). I used a single prompt for both: “build Mario 1-1 in HTML/Canvas, playable” (I will provide the complete prompt below) ... and the 31B local model completely one-shotted it! Sure, it's a bit slower (2m53s vs 59s), but getting a fully playable game locally on a laptop was impossible a year ago. The cloud moat (Claude, GPT, Gemini) is shrinking fast. The secret sauce here is the AtomicChat app. Combined with TurboQuant under the hood, it perfectly optimizes these heavy open-source models so they can conquer REAL coding tasks on your Mac, not just benchmarks 👊 Why local is the future: → ZERO API limits → 100% private and offline → 24 tok/sec on a 31B model! → ZERO subscription costs I've pasted the prompt below 🧵↓
English
27
45
186
24.6K
Victor | Structural Alpha
Victor | Structural Alpha@MarginOfAlpha·
@DataChaz @VelobaseX fork and win assumes the hard part was the billing logic. their 8 figures came from being early in a window that closed, not from prettier invoices. i've got a repo full of other people's moats sitting at zero stars.
English
1
0
0
23
Charly Wargnier
Charly Wargnier@DataChaz·
HUGE WIN FOR OPEN-SOURCE If you want to see what an 8-figure ARR stack looks like, @VelobaseX just made theirs public. You get usage-based billing, affiliates, even server-side attribution for Ads & 𝕏 👀 Monetization is the new moat. Fork and win.
Velobase@VelobaseX

Everyone can build an app now. Almost no one makes a dollar from it. We went from the same problem to 8-figure ARR. The secret wasn't the product — it was the infrastructure behind it. Today we're open-sourcing the whole stack! 🧵

English
6
25
46
6.8K
Charly Wargnier
Charly Wargnier@DataChaz·
repo link → github.com/decolua/9router Shoutout to Decolua for building this and making it open-source for the community! 🤗 Don't forget to drop a ⭐️!
Charly Wargnier tweet media
English
0
3
11
1.4K
Charly Wargnier
Charly Wargnier@DataChaz·
🚨 Jira is the central brain for millions of devs. This new integration with Cursor is about to change how they ship code FOREVER 💥 Imagine never having to leave your IDE to update a ticket again. (now read that again) Here's the cheat code: > The new Teamwork Graph CLI handles it all directly from your terminal > Just assign a task to the cloud agent and let it work > It instantly syncs all back to Jira when the PR is ready for review man I SO wish I had this kind of superpower back when Jira was my entire life 😅 announcement below ↓
Tamar Yehoshua@TYehoshua

We continue to make @Atlassian tools the place for AI-driven software teams. Teams in Jira can now directly assign any work item to @cursor_ai and a cloud agent will get to work. And it works both ways. If you're working with Cursor in your IDE or terminal, you can keep your teams fully up to date in Jira with our native Teamwork Graph CLI. Context is instantly shared, so there is less tool switching and more time in flow. It's now available with every paid Jira subscription. More here: atlassian.com/blog/company-n…

English
11
38
65
15.5K