🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Vibe Coding, 2 devs will be like:

What's going on here? With any other leading code gen tool it would have taken 1-2 prompts to complete this simple game.. What am I missing?

Our team at Qodo tested OpenAI's new GPT-4.1 on a real-world coding task: providing code review for actual pull requests. After running it through 200 PRs, here’s what stood out: GPT 4.1 was better than Claude 3.7 Sonnet 55% of the time. - It provided more focused,

>>> A window that can hold the entire contents of “War and Peace,” tenfold? It’s the kind of headline that lights up conference stages and Twitter feeds. But as the dust settled and I reflected—not just as an AI enthusiast, but as someone deploying these systems in the real

#Automation vs. #Autonomy - and why the difference matters more than ever for developers, and how this relates to #VibeCoding Vibe Coding means different things to different people. Often, it is used informally by developers to describe a relaxed, creative, and intuitive coding

Vibe Coding today often feels like punching in the dark. We chase bugs. We guess. We deploy and pray. This isn’t engineering. I guess this is why this video circulated in social In Vibe Coding, we “feel” our way through a codebase, guided more by intuition than understanding.

Phase #1: VS Code extensions Phase #2: VS Code forks Phase #3: ?

Wow, look at this. Karpathy, who coined the term #VibeCoding, now shares his reflections: "Code I actually and professionally care about, in contrast to vibe code" Vibe Coding isn't enough, and it might even be dangerous! But it will evolve. Bringing in relevant, essential

Let’s clarify Agent-to-Agent (#A2A) vs. Model Context Protocol (#MCP) – In their current implementations, I believe that these are two concepts that advocate for completely two different agent architectures. MCP is built for having a `𝒔𝒖𝒑𝒆𝒓 𝒂𝒈𝒆𝒏𝒕` that uplifts its

Qodo Gen now supports Claude Sonnet 4 & Opus 4! This latest model takes its coding capabilities to the next level — exceling at coding and complex problem-solving, powering frontier agent products. Shoutout to our friends at Anthropic for continuing to push the envelope 👏

Super interesting. Different coding tasks do have different requirements from models, e.g. understanding hunks According to Qodo's Pull Request Benchmark, Sonnet 4.0 is an improvement over Sonnet 3.7, but, seems to be inferior to both GPT-4.1 and especially Gemini-pro 2.5

I knew this day will come! AI keyboard from Traceloop Found at #aiengineering conf, but soon everywhere ? swyx, I hope you you got one ;)

The new Gemini-pro-06-05 model includes an option to control the thinking token budget, making it viable for latency-constrained environments. But how does this new model apply to real-life problems? Well, it depends dramatically on how much it's allowed to think. 🤔 Using the

💥 NVIDIA CEO Jensen Huang just spotlighted Qodo’s context-aware engine live on stage — built with NVIDIA’s software stack and designed to deliver high-precision code retrieval at enterprise scale

Asimov's "Three" Laws of Robotics seem more relevant than ever! FYI, I tried 4 times to make ChatGPT generate the complete Zeroth Law correctly, and could not - either it cut the end of it, or didn't finish it nicely - Why? 🤔

“I vibe coded all the way to prod.” — Every PM in 2025 with GenAI access