Edra exits stealth with $30M from Sequoia, automating enterprise workflows by watching employees work
Mar 19, 2026 with Eugen Alpeza
Key Points
- Edra exits stealth with $30M Series A led by Sequoia Capital, solving enterprise automation by observing employee workflows across ServiceNow, Jira, and Salesforce instead of requiring explicit process documentation.
- The company closes deals in one week with proof-of-concept datasets rather than three- to six-month discovery cycles, with early customers including Asos, Cushman Wakefield, and HubSpot.
- Edra deploys expensive frontier models for initial workflow discovery, then shifts to lighter models for execution, trading upfront compute cost for downstream efficiency at scale.
Summary
Edra raised $30M in Series A funding led by Sequoia Capital, with participation from Y Combinator and prior seed investors. The company observes employee workflows across enterprise systems like ServiceNow, Jira, Outlook, Salesforce, and Zendesk, then documents and automates those processes without requiring explicit instructions.
Most AI models can handle enterprise work tasks, but companies struggle to specify what they want automated because workflows are often undocumented or poorly understood. Edra's approach is to watch what employees actually do and build automation from that behavior.
Co-founder and CEO Eugen Alpeza positions speed as a competitive advantage. Instead of a three- to six-month discovery process, Edra offers a one-week proof of concept. Provide one static dataset and the system surfaces operational insights the company did not know about. Named customers include Asos, Cushman Wakefield, and HubSpot, where process complexity and scale make observation-based automation especially valuable.
Edra uses frontier models for initial workflow discovery and documentation, where reasoning power matters most. Once workflows are understood, the platform shifts to lighter models for execution, which are sufficient for instruction-following tasks. This splits the compute cost between expensive reasoning at the start and efficient execution later.
Alpeza spent over a decade in process automation, which supplied the network to close early enterprise deals while in stealth. The company exits stealth with customer wins already in hand, suggesting validation at scale before public launch.