This AI Founder Says Prompting AI Agents Is a Lot Like Managing People


The very best managers of AI agents could also be individuals who have already managed people, one AI founder stated.

Ivan Burazin, the cofounder and CEO of New York-based AI agency Daytona, has a workforce of about 30 folks throughout groups within the US and Europe. He informed Enterprise Insider that in latest months, he is pushed his employees to construct extra AI brokers; his 16 engineers every run a median of 5 brokers, and nobody on the firm writes code anymore.

That is introduced one factor to mild for him: Workers who’ve managed others are approach higher at prompting AI brokers to get precisely what they need than these with no people-management expertise.

“We have discovered that individuals who handle folks and perceive find out how to talk what they need — inputs versus outputs — they are much higher at operating brokers,” he stated.

Then again, an engineer who’s not managed anybody, even when they’re nice at their job, is wired like a person contributor, Ivan stated. He stated they implicitly anticipate that individuals — and brokers — will perceive what they’re asking for, and get harassed and annoyed when it would not pan out.

“And they also by no means take into consideration find out how to talk, ‘That is what I need. That is how I anticipate you to do it, or what I feel the end result needs to be,'” he added.

That stated, Burazin would not assist the concept of individuals managers alone. He stated managers with no technical information — corresponding to in advertising or in engineering — are usually not helpful, and everybody needs to be getting their arms soiled with precise work.

Burazin and his cofounders, Vedran Jukić and Goran Draganić, based Daytona in 2023, which offers safe sandbox infrastructure for corporations to run AI brokers and AI-generated code in isolated environments. The corporate lately raised $48 million in a Collection B funding spherical, in keeping with Crunchbase.

Burazin’s feedback align with how managerial roles in tech corporations have modified lately.

After an period dubbed “The Great Flattening,” by which corporations like Meta, Amazon, and Google laid off center administration for effectivity, more tech leaders are actually juggling coding and product improvement alongside employees administration.

Burazin says he helps his employees operating a number of AI brokers, as a result of it helps him preserve his workforce small. Hiring at Daytona is “very gradual,” he stated, as a result of every new rent provides a layer of complexity.

“Brokers assist with that as a result of you possibly can have smaller human groups, therefore the communication is smaller, and you may proceed to be quick,” he stated.





Source link