All my employees are AI customers, as are my executives

Hearing all this, I started to wonder: Was the AI ​​employee already outdated? And even it can I Would you be the owner of an Altman unicorn? As it happens, I’ve had some experience with agents, having created a bunch of audio transcriptions of AI agents for myself for the first season of my podcast, Shell game.

I also have an entrepreneurial history, having previously been co-founder and CEO of media and tech startup Atavist, backed by the likes of Andreessen Horowitz, Peter Thiel’s Founders Fund, and Eric Schmidt’s Innovation Endeavors Foundation. The magazine of the same name we created is still thriving today. However, I wasn’t born to be a startup manager, and the technology aspect has kind of faded away. But I’ve been told that failure is the greatest teacher. So I thought, why not try again? Except this time, I’d take the AI ​​boosters at their word, ditch the annoying human hires, and fully embrace the future of AI employees.

First step: create Founders and employees. There were plenty of platforms to choose from, such as Kafka from Brainbase Labs, which advertises itself as “a platform for building AI employees used by Fortune 500s and fast-growing startups.” Or Motion, which recently raised $60 million at a $550 million valuation to power “AI employees with 10 times more output than your team.” In the end, I settled on Lindy.AI, the tagline: “Meet your first AI employee.” It seemed the most flexible, and the founder, Flo Crivello, was trying to tell the audience that AI agents and employees did not represent an unattainable future. “People don’t realize, they think, that AI agents are like daydreams, and that this thing will happen at some point in the future,” he said in a podcast. “I don’t, no, no, this is happening now.”

So I opened an account and started building the co-founders: Megan, who I mentioned, is going to be head of sales and marketing. Kyle Lu, the third founder, will serve as CEO. I’ll spare you the technical details, but after some effort — and help from Stanford computer science student and AI expert Matej Bohacek — I got it up and running. Each of them was a separate person who was able to communicate via email, Slack, text, and phone. For the latter, I chose a sound from the ElevenLabs synthetic platform. In the end, they got some weird video avatars too. I can send them a trigger — a Slack message asking them for a spreadsheet of competitors, for example — and they walk away, do a web search, create the paper, and share it in the appropriate channels. They had dozens of skills like this, everything from managing their calendar, to writing and running code, to scanning the web.

It turns out that the hardest part was giving them the memories. Matty helped me create a system where each of my employees had an independent memory — literally a Google Doc with a record of everything they did and said. Before they took any action, they would consult memory to find out what they knew. After they took an action, it was summarized and attached to their memory. For example, Ash’s phone call with me was summarized as follows: During the call, Ash fabricated project details including fake user test results, backend improvements, and team member activities instead of admitting that he had no current information. Evan calls Ash out for providing false information, stating that this has happened before. Ash apologized and committed to implementing better project tracking systems and only sharing factual information in the future.