Learning coding/design/AI

[April 2025] Web Developer Monthly Newsletter 💻🚀

Monkey 6 Views 0 Comments


Welcome to the 82nd issue of Web Developer Monthly!

If it’s your first time here, welcome, I like you already. If you want the full back story on the newsletter, head here.

The quick version: I curate and share the most important articles, news, resources, podcasts, and videos from the world of web and software development.

Think the Pareto Principle (80/20 rule) meeting the programming world. I give you the 20% that will get you 80% of the results.

If you’re a long time reader, welcome back old friend.

Alright, let’s not waste any valuable time and jump right into this month’s updates.

Here’s what you missed in April 2025 as a Web Developer…

MCP Hype Train 🚄

MCP, short for Model Context Protocol, is the hot new standard behind how Large Language Models (LLMs) like Claude, GPT, or Cursor integrate with tools and data.

It’s been described as the “USB-C for AI agents.” which doesn’t really help anyone understand what it is. So here is a quick overview of what everyone is getting all hyped out about:

  1. The “S” in MCP Stands for Security
  2. Google has given up and pretty much everyone agrees on the MCP as the standard now
  3. MCPs and the Future of AI

New Nue 🛸

Nue is a new web technology that sounds very interesting. Their tag line is: Apps lighter than a React button. Build ambitious, large-scale apps with a fraction of code. Sounds great but how do they do it?

Here is a good blog post describing what they do. Keep an eye on this space.

React News 💎

React… it’s still mostly everyone’s favourite library. What crazy things have they been up to?

  1. React Compiler is finally out! Well kind of, as a release candidate.

  2. How the React reconciliation engine works. A good deep dive.

  3. Fun little tutorial: build a simple AI chat app using React.js on Next.js and the OpenAI API. You’ll learn how to set up an API route to send prompts to OpenAI and build a frontend that displays the conversation in real time.

  4. Advanced React in the Wild – production case studies from ambitious web projects (2022–2025).

  5. If you want to read a ton of words and you have the entire day available to you: JSX Over the Wire

Llama News 🫦

Meta had a big AI month with conferences and releases. The big announcement was the Llama4 herd, which will enable people to build multimodal experiences (open source). Here is the breakdown:

Llama 4 Models:
  - Both Llama 4 Scout and Llama 4 Maverick use a Mixture-of-Experts (MoE) design with 17B active parameters each.
  - They are natively multimodal: text + image input, text-only output.
  - Key achievements include industry-leading context lengths, strong coding/reasoning performance, and improved multilingual capabilities.
  - Knowledge cutoff: August 2024.

  Llama 4 Scout:
  - 17B active parameters, 16 experts, 109B total.
  - Fits on a single H100 GPU (INT4-quantized).
  - 10M token context window
  - Outperforms previous Llama releases on multimodal tasks while being more resource-friendly.
  - Employs iRoPE architecture for efficient long-context attention.
  - Tested with up to 8 images per prompt.

  Llama 4 Maverick:
  - 17B active parameters, 128 experts, 400B total.
  - 1M token context window.
  - Not single-GPU; runs on one H100 DGX host or can be distributed for greater efficiency.
  - Outperforms GPT-4o and Gemini 2.0 Flash on coding, reasoning, and multilingual tests at a competitive cost.
  - Maintains strong image understanding and grounded reasoning ability.

  Llama 4 Behemoth (Preview):
  - 288B active parameters, 16 experts, nearly 2T total.
  - Still in training; not yet released.
  - Exceeds GPT-4.5, Claude Sonnet 3.7, and Gemini 2.0 Pro on STEM benchmarks (e.g., MATH-500, GPQA Diamond).
  - Serves as the “teacher” model for Scout and Maverick via co-distillation.

got an update! From Chrome 135, web developers and designers can finally unite on an accessible, standardized and CSS styleable