Trace raises $3M to solve the AI agent adoption problem in enterprise
YC-backed startup Trace builds 'manager' AI that orchestrates agents across Slack, email, and Airtable with knowledge graphs.
Trace, a London-based workflow orchestration startup from Y Combinator's 2025 summer cohort, has raised $3 million in seed funding to tackle the slow adoption of AI agents in enterprises. Co-founded by CEO Tim Cherkasov and CTO Artur Romanov, the company argues that agents from labs like OpenAI and Anthropic are like "brilliant interns" lacking the contextual understanding of complex corporate environments. Trace aims to be the "manager" that provides this missing layer by mapping a company's existing tools—such as email, Slack, and Airtable—into a unified knowledge graph. This foundational context allows users to assign high-level tasks (e.g., "develop our 2027 sales plan"), which Trace then breaks down into step-by-step workflows, intelligently delegating between AI agents and human workers.
The technical approach hinges on what Romanov calls a shift from "prompt engineering to context engineering." When Trace's system invokes an AI agent for a subtask, it automatically provides the specific data and background needed from the knowledge graph, eliminating the manual, delicate work of onboarding agents that currently blocks enterprise deployment. The $3M seed round was led by Y Combinator and includes Zeno Ventures, Goodwater Capital, and angel investors. However, Trace faces significant competition from Anthropic's new enterprise agent plugins and native agents being built into platforms like Atlassian's Jira. The startup's bet is that its agnostic, knowledge-graph-based orchestration will become the essential infrastructure layer for AI-first companies, providing superior context at the right time to make agentic workflows actually scalable.
- Raised $3M in seed funding from Y Combinator, Zeno Ventures, and Goodwater Capital to build enterprise AI agent orchestration.
- Builds knowledge graphs from tools like Slack, email, and Airtable to provide agents with critical operational context.
- Automates workflow creation and delegation for high-level tasks (e.g., 'design a microsite'), shifting from prompt to 'context engineering.'
Why It Matters
It automates the complex setup blocking enterprise AI agent deployment, potentially unlocking scalable, context-aware automation for business processes.