Simular is building desktop-level AI agents capable of controlling the cursor, keyboard, and interface of Mac OS and Windows PCs to complete multi-step tasks such as copying data into spreadsheets or extracting information from PDFs. Unlike browser-based agents, Simular’s technology interacts with the entire operating system, using a "neuro-symbolic" approach that lets an LLM explore solutions and then converts successful workflows into deterministic, auditable code. This method aims to curb hallucinations while giving end users transparent, reusable automations. The company released its 1.0 Mac OS agent and is collaborating with Microsoft through the Windows 365 for Agents program to launch a Windows version. Early beta users include a car dealership automating VIN searches and homeowners’ associations parsing contracts, while its open-source community has built sales, marketing, and content-creation automations. Founded by former Google DeepMind researchers Ang Li and Jiachen Yang, Simular has raised approximately $27 million to date, including a newly closed $21.5 million Series A and an earlier $5 million seed round.

Total raised

$22M

Funding rounds

1

Latest round

Series A

Latest activity

Dec 2025