Bourne raises $50M in pursuit of the 10x sales rep

Today we announced a new round of funding with participation from Bessemer Venture Partners, Y Combinator, and Uncorrelated Ventures, bringing our total funding to over $50 million.

What we are

Bourne is a research lab working on human-agent collaboration in high-judgement domains.

The consensus playbook for AI value says to go the other way: start where the work is mostly intelligence, tasks with complex but learnable rules that can be automated end to end, and defer the judgement-heavy work until the data compounds. It is sound advice, and it is why most AI companies are crowding into the same low-judgement corner.

But most of the economic value does not live in that corner. It lives in high-judgement domains, where the cost of a wrong call is real and the right call depends on context no model picks up for free. No lab has figured out how to reach full efficiency of AI there. Enterprise deployments in these domains fail far more often than they work: the rollout stalls, every output gets reviewed into oblivion, and the team quietly goes back to doing the work by hand.

We do not believe that is a model-capability gap the next release will close. It requires new methods of agent design and development, built around humans and agents as one system: agents that know which calls are theirs, which belong to a person, and how to get better at drawing that line.

Sales is the first domain. Not because it is the easiest, but because it is the clearest case of the problem we care about. A great rep makes hundreds of judgement calls a week, and almost everything around those calls, the research, the follow-ups, the writing, the record-keeping, is work that machines should be doing by now.

That is the premise behind the 10x sales rep. We build and operate AI revenue systems so that one rep can carry what used to take a team, and a new rep can ramp in weeks instead of quarters.

What we learned getting here

We earned this thesis the slow way. We built HockeyStack, and over the last few years grew it to more than 300 enterprise customers, including Fortune 100 revenue teams.

Building HockeyStack put us in the room with the best revenue leaders in the world. We watched how they run pipeline, how they coach, how they decide which deals are real, and where their best reps actually spend their judgement. The pattern was consistent: the highest-leverage moments in a deal are human, and almost everything around those moments is work the rep should not be doing. That observation is what Bourne exists to act on.

People are the differentiator

The point of this work is not to remove people from selling. It is the opposite. In the age of AI, the biggest differentiator will be people. The systems we build execute around the clock so that the humans can spend their time on the two things only humans can do: building relationships and closing.

Human-agent collaboration is a research problem, and we treat it like one. Where should the agent decide alone? Where should it bring the rep in? How does institutional knowledge move from the heads of the best people into the system without flattening into a script? These questions do not have settled answers. Our job is to settle them in production, with real quota on the line.

What’s next

The new capital goes into research, and we are hiring exceptional researchers to do it. If you want to work on high-judgement human-agent collaboration with real revenue as the scoreboard, email me at bugra@usebourne.com.

To our customers, who pushed us to build this, and to Bessemer, Y Combinator, and Uncorrelated, who backed the bet: thank you. We are just getting started.