Architected for Scale and Compliance
Explore the engine powering real-time employee timelines, native desktop captures, and high-performance analytical storage.
Native macOS Engine
Our lightweight daemon client utilizes compiled AppleScript hooks to query the frontmost window name and queries the native IOHIDSystem kernel registers directly to audit precise mouse click, keystroke activity, and active idle seconds.
- <0.5% CPU performance overhead
- Swift-compiled image processing filters
- Offline caching log recovery queue
ClickHouse time-series scale
Activity logs are written as time-series logs in a column-oriented ClickHouse database engine. This allows company supervisors to scan millions of system focus state pings and generate aggregated timeline distributions on demand.
- Columns compressed using ZSTD filters
- Under 50ms aggregate response query
- Batch insert routes reducing writing locks
Presigned S3 pipelines
Avoid passing high-bandwidth file chunks through your application server. The agent asks the Go backend for a secure S3-compatible presigned URL and uploads the blurred JPEG directly to MinIO or DigitalOcean Spaces storage.
- Secure signed PUT links with 15m expiration
- Signed GET links ensuring auth restriction
- Zero server-side file buffering required
Timeline Distribution Builder
Select or toggle the active work blocks below to simulate how the classification engine groups categories and updates the daily productivity scores.
Productivity Score
Aggregated distributions based on selected mock logs.