It has been less than three years since Shawn "swyx" Wang coined the term AI engineer in an open letter to the community, but this week thousands of engineers are converging on San Francisco for the now-annual AI Engineer World’s Fair.
As he put it, AI engineers don’t need doctorates and the ability to create LLMs. Rather, they do the important work of integrating machine learning capabilities into current software effectively. “When it comes to shipping AI products, you want engineers, not researchers,” he suggested.
While the latest AI models from the likes of Anthropic, OpenAI, and Google tend to grab the consumer press headlines, the task of integrating these new capabilities into workable, sellable software is arguably more important. And adding APIs without breaking the code is a profession that appeals to many software engineers.
The concept spawned a San Francisco summit, and in 2023 there were 500 software engineers discussing the ins and outs of using this relatively new technology, with many more attendee applications than there were places.
The demand was so great, in 2024 it was renamed the AI Engineer World’s Fair, and all the big software companies were there to show off their latest advances and advise on how to use them in applications. Not to be left out, Nvidia’s CEO Jensen Huang made a couple of surprise appearances to point out his hardware was ready for this.
In 2026 the conference has now spread over four continents, with events in London, San Francisco, and New York, as well as partner conferences in Miami, Singapore, Melbourne, Paris, and Sydney. It looks set to grow further, given the interest in the topic, and software engineers who once saw their jobs as threatened by AI are now embracing it for the roles it will provide in the future.
