Pinned Tweet
🎉 I’m excited to announce that I’ve just finished my Laravel for Beginners course!
After working on this for the past 6 months, I’m super proud to offer you a well-planned, clean, polished, and carefully edited Laravel course, perfect for beginners! 🙌
thecodeholic.com/p/laravel-1…
Key Features:
📼 24 hours Full HD video
📚 300+ videos and 37 modules
🏆 Quizzes and challenges in almost every module
✍️ Written notes, right under the videos
🎧 English captions
💻 Downloadable videos
🏅 Certificate of completion
🌐 Step-by-step guide on deploying your project to a custom domain on a VPS using GitHub Actions
😎 No extra JS or CSS frameworks
🤝 Access to a private Discord group
🔓 Lifetime access
💻Fully working source code
#laravel #laravelcourse #phplaravel #thecodeholic
The Codeholic retweeted
BREAKING NEWS!
The engineer who built Claude Code from scratch just released a 28-minute video that's pure gold: how to write prompts that deliver ridiculously good results.
I've seen $300 courses that don't even come close to what he explains in the first 10 minutes.
CLAUDE.md files, memory shortcuts, parallel sessions, and prompting patterns that almost no one uses...
All in one single video. Completely free. No fluff.
Whether you're a developer, just starting, or have already been using Claude for months, this changes the game for you starting today.
Save. Watch. Share.
Join us tomorrow to celebrate one year of Laravel Cloud.
We'll be streaming all day with:
◆ @taylorotwell
◆ @JohnONolan
◆ @jessarcher
◆ @LeahTCodes
◆ @christophrumpel
◆ @ericlbarnes
Sign up for a chance to have your app featured, win swag, Cloud credits, and more.
cloud.laravel.com/one-year-o…
I think Codex app and GPT5.3 the best coding model right now. Really smart, understands what I want and does really good job and with current double Credits I don't even think using Claude code anymore.
Getting my platform ready to be launched in next week, making final tweaks and tests...
thecodeholic.com will be much better.
I really did not like the code Claude Code wrote so I thought a little bit and came up to a simpler solution.
It aknowledged that my approach was better...
P.S.
I have rules to not agree with me (Like AI does in most cases 🙂) and be honest and critical.
OpenClaw tip (save money):
• Big chats = big cost
• Use 1 group per topic
• When done, start a new session
• Write 3 lines of summary so you don’t forget
#openclaw
🚀 Big update!
I’ve built my own online courses platform and I’m starting the migration away from Teachable.
Teachable worked… but not smoothly:
• payment issues for students
• occasional downtime
• admin experience wasn’t great
So I decided to use my 14 years of software development experience and build everything from scratch.
✨ What’s coming:
• all my existing courses on the new platform
• new courses very soon
• AI-based video summarization and more cool features
👀 Early access
If you’re already enrolled in my courses and want to test the new platform before the full migration, DM me on X. I’d love your feedback.
Can’t wait for you to try it 🙌
thecodeholic.com
I tested a multi-agent AI setup (OpenClaw + Anthropic models).
Cloud Opus delegated work to Cloud Sonnet, ran FFmpeg edits, replanned, retried…
4 hours later: $50 spent.
Zero usable output.
Multi-agent ≠ productivity.
AI didn’t fail.
Unbounded autonomy did.
Without strict budgets, stop conditions, and success criteria,
multi-agent systems are demos — not production tools.
I tested a multi-agent AI setup (OpenClaw + Anthropic models).
CloudOpus delegated work to CloudSonnet, ran FFmpeg edits, replanned, retried…
4 hours later: $50 spent.
Zero usable output.
Multi-agent ≠ productivity.
Agent-to-agent delegation is a hidden cost multiplier.
Every handoff:
• resends context
• re-parses instructions
• generates verbose output
Two agents “collaborating” can cost more than one agent doing real work.














