Role overview
Build the real-time services behind a voice-first assistant for dictation, meetings, and everyday work. This backend-leaning full-stack role spans streaming speech recognition, AI features, and production operations, with particular care for reliability and the handling of sensitive workplace conversations.
Responsibilities
- Design and operate real-time services for speech-to-text, long-running meeting and notes workflows, and web, desktop, and mobile clients.
- Develop AI-powered capabilities such as summaries and action items, and connect them with calendars and meeting platforms.
- Create evaluations and live quality monitoring to track how features perform for users.
- Own production reliability through observability, security, and sound data-handling practices; participate in the on-call rotation.
- Take features from idea to release in collaboration with product, design, and engineering, contributing to frontend work when needed.
Requirements
- Strong backend engineering skills and experience operating production systems, including APIs, cloud infrastructure, storage, and asynchronous processing.
- Experience building reliable real-time systems, such as streaming services, WebSockets, and background workers, with graceful failure recovery.
- Hands-on experience shipping LLM-powered features and working with prompts, tools, retrieval, and evaluations.
- Product judgment and a record of independent, end-to-end ownership, with the ability to work across the stack.
- Clear communication and sound prioritization in a small, fast-moving team.
Nice to have
- Experience with speech or audio systems, calendar or meeting integrations, privacy-sensitive data, or mobile backends.
Benefits and work setup
- Remote-first, with the ability to work across GMT-5 to GMT+8.
- Offices are also available in several cities.
- Support includes annual professional-development and coworking stipends.
- A yearly team offsite.
- A discretionary stipend for meeting colleagues in person.