About the Role
Join a small, fast-moving engineering team building voice AI agents for customer-facing workflows. You will own projects end to end, working across applied AI, real-time audio, and distributed systems to solve practical customer problems.
What You'll Do
Build reliable voice AI systems, including retrieval-augmented generation, dynamic model routing, and testing transcription services before production rollout.
Design and operate a distributed call scheduler that manages concurrency, race conditions, time zones, retries, and provider limits.
Extend agents across voice, messaging, and email, with shared memory and state for asynchronous and real-time conversations.
Develop memory and reasoning systems that preserve context throughout customer workflows.
Build AI-assisted tools for configuring agents, debugging issues, and connecting to business systems such as CRMs, calendars, and email.
Improve tracing and resilience so failures across calls, workers, and providers can be diagnosed and handled effectively.
What We're Looking For
Production experience building and owning distributed product systems, and a track record of shipping projects end to end.
Experience with Python, FastAPI, or TypeScript in production, plus PostgreSQL, Redis, or comparable data stores.
Background in real-time systems, audio processing, or voice infrastructure, and experience with concurrency, load balancing, or fault tolerance.
Experience shipping LLM, AI model, or agentic systems, and working with AWS, GCP, or Azure.
Familiarity with workflow orchestration or message queues, observability tools, and full-stack development is useful.
Technologies include LiveKit, Bun, Node.js, Fastify, React, Temporal, and OpenTelemetry.
Strong software fundamentals, independent problem-solving, and a team-oriented focus on customer and business outcomes.
Compensation & Benefits
€85,000 to €160,000 annually, plus equity. Visa sponsorship is available. Relocation support is available.
Location
On-site in Berlin, Germany.