See how one developer used AI to collapse a decade-long goal into a single quarter. Also: Nvidia eGPUs on Arm Macs and the state of CSS.
Nearly 30% of code is now AI-generated. Median site ships 309 kB CSS with mixed modern/legacy code. TypeScript and Vite dominate JS. Plus: why continuous integration needs to fail to provide value.

View in browser | Past Issue | Subscribe / Unsubscribe

SitePoint Source

Welcome, Developers! 👋

AI tools are redefining development speed, as shown by an eight-year AI project accomplished in three months. Meanwhile, Apple's approval for Nvidia eGPUs on Arm Macs shakes up hardware compatibility. Uncover how employers leverage personal data in salary talks. Dive into these stories and explore tools for AI, NLP, and fine-tuning strategies.

From our sponsor: Superhuman

Find out why 200K+ engineers read The Code every day

Staying behind on tech trends can be a career killer.

​

But let’s face it, no one has hours to spare trying to stay updated.

​

That’s why over 200k+ engineers at companies like Google, Meta, and Apple read The Code newsletter.


Here's what you get:

  • AI coding techniques used by top engineers 

  • Tools and workflows that cut your coding time in half

  • Tech insights that keep you 6 months ahead
Join 200,000+ engineers

🔖 The Reading Room

Articles we have hand-picked for you:

The threat is comfortable drift toward not understanding what you're doing

This opinion piece warns of complacency in tech comprehension, driving home the risk of developers losing touch with the why behind their actions. It explores the subtle decline in understanding as automation and advanced tools take center stage. This drift could impact the quality of development and innovation, emphasizing the need for developers to stay engaged actively.

By Ergosphere →

Eight years of wanting, three months of building with AI

A developer shares their journey of creating with AI, condensing an eight-year goal into a mere three months of work. The piece delves into the efficiencies achieved using AI technologies, like Syntaqlite AI. This rapid development cycle highlights how AI can accelerate project timelines, allowing developers to pursue ambitious innovations swiftly.
By Lalit Mehta →

Employers use your personal data to figure out the lowest salary you'll accept

An investigation reveals how employers exploit personal data to determine the minimum salary threshold employees might accept. This practice raises ethical questions and underscores the importance of privacy in modern workplaces. For tech workers, this could mean negotiating from a position of vulnerability unless privacy safeguards improve.

By MarketWatch →

Running Gemma 4 locally with LM Studio's new headless CLI and Claude Code

This tutorial provides a comprehensive guide to running Gemma 4 locally with new tools like LM Studio's headless CLI and Claude Code. The focus is on maximizing setup efficiency and customization potential. This facilitates developers with more control over AI tools without reliance on cloud-based solutions, enhancing accessibility and bespoke development environments.

By GeorgeLiu AI →

Apple approves driver that lets Nvidia eGPUs work with Arm Macs

Apple has green-lit a driver enabling Nvidia eGPU compatibility with Arm-based Macs, revolutionizing peripheral options for developers needing extra graphics power. This move removes performance barriers previously faced by Mac users and provides new opportunities for graphics-intensive applications, gaming, and development tasks on Apple hardware.

By Tinygrad →

⏳ Back in Time

Most clicks from last newsletter:

🔗 The Link Lounge 

Unordered finds from around the web:

Find something cool? You can send us links to feature here via email.

🧰 The Toolbox

Tools and products we're excited about today:

Nimbalyst

A free, local, visual workspace for building with Claude Code and Codex. Manage 10+ parallel agent sessions from a kanban board, review live diffs inline, edit markdown, mockups, diagrams, and code with your agent. Available for macOS, Windows, and Linux.

Learn more →

Caveman: Why use many token when few token do trick

A minimalist natural language processing tool that focuses on efficient token usage, simplifying complex language tasks with fewer computational resources.

Learn more →

Show HN: I built a tiny LLM to demystify how language models work

A compact tool providing educational insights on language models, aimed at demystifying the inner workings and applications of these AI systems.

Learn more →

R6410418/Jackrong-llm-finetuning-guide

A comprehensive guide on fine-tuning language learning models, featuring Jupyter Notebook support to streamline the customization process.

Learn more →

Find out why 200K+ engineers read The Code every day

Staying behind on tech trends can be a career killer.

​

But let’s face it, no one has hours to spare trying to stay updated.

​

That’s why over 200k+ engineers at companies like Google, Meta, and Apple read The Code newsletter.


Here's what you get:

  • AI coding techniques used by top engineers 

  • Tools and workflows that cut your coding time in half

  • Tech insights that keep you 6 months ahead

Join 200,000+ engineers →

🎤 Your Voice

Your feedback shapes what comes next! We read every email, so simply hit reply and tell us what's on your mind.

71 Balmain Street

Cremorne

Vic

3121

Australia

You received this email because you signed up on our website or made a purchase from us.

You can unsubscribe using this link.