I just published a two-volume book series called "The AI-Augmented Developer's Rust Series."
The core thesis: AI coding assistants have fundamentally lowered the barrier to using Rust. You no longer need to be a Rust expert to be productive—the AI handles the syntax, and the compiler handles correctness.
But there's a problem with AI-assisted development that nobody talks about enough: errors compound. You prompt, AI generates, you accept, you prompt again. Each small mistake builds on the last. And because things are running 100 miles per hour, the compound is fast. In dynamic languages, you often don't discover the accumulated mess until you're deep in a codebase you can't easily recover from.
Rust breaks this cycle - at least partially. The compiler forces you to confront issues immediately—before they compound. Or rather it forces AI to confront these issues upfront. Issues are fixed as you go, not six months later in a production incident.
The other insight: Rust covers the full stack. Yew for WebAssembly frontends, Axum for backends. One language, one mental model, shared types between client and server and strong typing although through to your database!
Would love the hear from you what tech stack you've had the most success with using AI? And with the time you save on coding, are you experimenting, figuring out what works best for you or how are you using the time you gain?