Marmelab founder and CEO, passionate about web technologies, agile, sustainability, leadership, and open-source. Lead developer of react-admin, founder of GreenFrame.io, and regular speaker at tech conferences.
Developers often try to eliminate errors from AI agents. But science shows that being wrong — within reason — is what makes intelligence adaptive.
A new breed of developers is emerging. They ship production code daily but can't write a for loop. They're not imposters – they're the future of programming.
Do we still need developers where a business analyst can build an app with Claude Code?
We built a proof-of-concept of a new mobile OS based on a LLM, where the notion of application disappears and the UI adapts to the current task. Is it the future of computer UX?
Various studies highlight the significant environmental footprint of Large Language Models (LLMs). Those concerned about climate change might consider avoiding AI for this reason. But is that the right choice? Let's examine the facts.
While AI agents can be built in days, it takes weeks to find the best configuration to optimize cost, speed, relevancy, etc. Eval tools and frameworks are the key ingredients for this optimization phase.
LLMs share similarities with CPUs but bring unique challenges like slower speeds and probabilistic behavior. To effectively use LLMs, we need new programming approaches, specialized developers, and adjustments in our educational system.
Mastering Digital Craftsmanship
© 2025 Marmelab
