Product updates, engineering deep-dives, and thoughts on the future of voice-first computing.
When most companies start with cloud-only, we decided to build local AI from day one. Here's why privacy and reliability drove our architecture decisions.
A deep dive into how Velocity detects which app you're using and formats your text accordingly—from emails to code comments.
Real-time transcription needs to feel instant. Here's how we used Cloudflare Workers and streaming to achieve sub-300ms latency.
Get product updates, engineering insights, and voice-first tips delivered to your inbox. No spam, unsubscribe anytime.