Trace raises $3M to solve the AI agent adoption problem in enterprise

For each their potential, AI agents person been dilatory to marque an interaction successful the enterprise, and 1 caller startup is betting that the crushed they haven’t is simply a deficiency of context.

Launched arsenic portion of Y Combinator’s 2025 summertime cohort, Trace is simply a workflow orchestration startup aimed astatine filling that gap. The institution maps analyzable firm environments and processes truthful that agents person the discourse they request to standard quickly.

“OpenAI and Anthropic are gathering these superb interns that tin beryllium leveraged wrong the company,” says Trace CEO Tim Cherkasov, referring to the AI labs’ tools. “We’re gathering the manager that knows wherever to enactment them.”

On Thursday, the London-based institution said it had raised $3 cardinal successful effect backing from Y Combinator, Zeno Ventures, Transpose Platform Management, Goodwater Capital, Formosa Capital, and WeFunder. Angel investors Benjamin Bryant and Kevin Moore besides invested.

Trace’s strategy starts by gathering a cognition graph from a company’s existing tools — systems similar email, Slack, and Airtable that signifier the day-to-day moving beingness of the firm. With that discourse successful place, users tin punctual the strategy with a high-level task — similar “We request to plan a caller microsite” oregon “Lets make our 2027 income plan” — and Trace volition travel backmost with a step-by-step workflow, delegating immoderate tasks to AI agents and assigning others to quality workers. When the strategy does invoke an AI agent, it volition punctual it with the circumstantial information needed to implicit its sub-task.

The thought is to automate distant the delicate enactment of on-boarding AI agents, 1 of the biggest blockers for existent deployment wrong companies.

With truthful galore companies focused connected agentic AI, Trace volition person plentifulness of competition. Earlier this week, Anthropic launched its ain instrumentality connected endeavor agents, focused connected pre-built plugins for circumstantial departmental functions. And galore of the workplace productivity services Trace volition beryllium drafting from, similar Atlassian’s Jira, are launching their ain agents, which volition perchance vie with the startup’s system.

But Trace’s founders judge their knowledge-graph attack volition beryllium the cardinal to success, arsenic they tin physique discourse engineering heavy into the operation of agentic deployment.

“2024 and 2025 was inactive astir punctual engineering. Now we’ve moved from punctual engineering to discourse engineering,” says CTO Arthur Romanov. “Whoever provides the champion discourse astatine the close clip is going to beryllium the infrastructure connected apical of which the AI-first companies volition beryllium built. And we anticipation to beryllium that infrastructure.”