Nun offers a platform that enables developers to automate legacy Windows desktop applications by creating software agents that interact with the UI through secure remote desktop sessions and computer‑vision detection. Automation logic is written in standard languages and invoked via a REST API, returning structured JSON while the service manages sessions, security, auditability, and observability.
Funding
Funding not disclosed
Founders
Product
Problem
Many enterprise workflows still depend on legacy Windows desktop applications that were designed for manual operation and lack accessible APIs. This makes it difficult to automate repetitive tasks, leading to manual data entry errors, slow processing times, and high operational overhead.
Solution
Nen offers an infrastructure platform that enables developers to create, deploy, and scale software agents capable of interacting with any Windows desktop UI. The platform establishes secure remote desktop sessions over standard protocols without requiring agents on the target machine, and it uses computer‑vision models to detect and manipulate UI elements. Developers write automation logic in familiar programming languages, package it as a workflow, and invoke it via a simple REST API, receiving structured JSON results. Nen handles session management, retry logic, and failure recovery, while providing enterprise‑grade security, auditability, and real‑time observability. The service abstracts the complexity of UI automation, allowing organizations to retain existing desktop tools while automating their use.
Target Audience
The primary customers are enterprise IT teams and developers in logistics, healthcare, and financial services who need to automate repetitive tasks on legacy Windows desktop applications without rewriting or replacing existing software.
Features
- Secure remote desktop connectivity using standard RDP/SSH protocols, with no client‑side installation and firewall traversal support
- Computer‑vision engine that visually identifies buttons, forms, and UI states to drive pixel‑accurate interactions on legacy applications
- Programmable automation defined in standard code (Python, JavaScript, etc.) without a proprietary DSL, enabling low‑latency execution alongside the desktop session
- On‑demand API endpoint that triggers workflows and returns results as structured JSON, suitable for integration with CI/CD pipelines, orchestration tools, or webhook handlers
- Workflow orchestration layer for reusable, versioned automation definitions that can be deployed across multiple endpoints
- Enterprise security controls including permissioned execution, role‑based access, audit logs, and end‑to‑end encryption of session data
- Built‑in observability and telemetry dashboards that capture execution traces, performance metrics, and error reports for continuous improvement
- Automatic session management with retry, state detection, and recovery mechanisms to handle UI redraws or unexpected interruptions