Hone, a San Francisco artificial intelligence startup, has raised a $60 million seed round led by Benchmark and Index Ventures to build what it calls Engines, AI systems meant to take ongoing responsibility for a business result rather than finish one task and stop.
The deal
Funding tracker FinSMEs reported the round values Hone at $285 million and that Elad Gil, Hanabi, Definition and Diffusion also took part. Hone said in an Oct. 8 blog post that it raised the money from Benchmark and Index and plans to use it to build out its Engines. The company also said it is hiring.
What the product is supposed to do
Most AI tools today work in episodes: a person types a prompt, the system responds, and the job ends. Hone says its Engines are different. A company assigns one a goal, such as improving customer retention, and the Engine learns the company's systems, history and people, decides what needs doing, and keeps working toward that goal over time. Hone says the Engines improve with feedback and that, because a whole team shares one Engine, a correction from one person helps everyone.
In its investment announcement, Index Ventures described the system exploring a company's software, studying past work and writing its own tools before being tested against earlier decisions and going live. Hone says the real test is financial: whether the Engine produces economic value.
Who is behind it
Chief executive Moritz Stephan studied computer science at Stanford and was an early employee at Cognition, the company behind the Devin coding agent, where Index says he worked on long-horizon agents and later served as chief of staff. Co-founder Oliver Brady was an early engineer at legal AI company Harvey and later helped lead enterprise work at Mercor. Co-founder Carlo Kobe, a Thiel Fellow, previously co-founded the startup Fizz with Stephan.
Why it matters locally
The round is another sign of how much venture money is flowing to San Francisco AI startups that promise to automate office work. Index said Hone is already working with AI companies and large enterprises, and has been hiring from firms such as Ramp, Midjourney and Decagon. The company has not publicly named its customers or said how many people it employs.