What you'll be doing

• Own and scale the voice pipeline for real-time transcription across devices.
• Evaluate and deploy new voice and transcription models in production.
• Build the agent harness for managing long-running jobs, tool calls, retries, and context retrieval.
• Extend APIs and integrations to bring Granola context into existing AI tools.
• Develop enterprise-grade trust features including permissions, data isolation, and admin controls.
• Manage Postgres performance, observability, and alerting as usage scales.

Requirements

• Experience owning backend systems in production and making trade-offs.
• Strong skills in TypeScript and Node.js.
• Deep knowledge of relational databases, particularly Postgres (query plans, indexing, zero-downtime migrations).
• Comfortable with cloud infrastructure as code, including AWS, Terraform, and AWS Lambda.
• Focus on security and anticipating failure modes.
• Strong product sense: understanding user needs and prioritizing shipping valuable features quickly.

Nice to haves

• Experience with speech or audio pipelines.
• Background in LLM infrastructure, evaluation systems, or streaming systems.

Perks and benefits

• Competitive salary and above-market equity.
• Relocation support to London.
• In-person work at a relaxed and flexible office in Old Street, including dress-down policy and dog-friendly environment.
• High-autonomy culture encouraging engineers to ship improvements and experiment with new models and tools.
• Collaborative, kind, and design-focused team environment with direct user engagement.