Software Engineer (Voice AI / Backend) at Clera
Germany
Posted
on-site · full_time · Visa sponsorship
Responsibilities:
- Design, implement, and operate distributed voice AI backend systems: call schedulers, RAG, LLM routing, and transcriber shadow-testing.
- Build reliable real-time audio pipelines and multi-channel agent flows (calls, WhatsApp, email) with shared memory and state.
- Implement memory and reasoning systems to preserve context across full customer interactions.
- Ensure observability and resilience: tracing, logging, root-cause analysis, retries, and fault-tolerance for calls and providers.
- Collaborate on architecture, push frontier models into production, and own projects end-to-end from design to deployment.
Requirements:
- 2–5 years production experience building and owning distributed systems, handling race conditions, load, and fault tolerance.
- Strong proficiency in Python (FastAPI) and/or TypeScript (Node.js/Bun/Fastify) with end-to-end shipping experience.
- Experience with real-time systems, audio processing, or voice infrastructure is a strong plus.
- Hands-on experience deploying LLMs or agentic systems in production.
- Familiarity with PostgreSQL, Redis, cloud platforms (AWS/GCP/Azure), workflow orchestration (Temporal or queues), and observability tools (OpenTelemetry/Honeycomb).
- Self-starter with bias toward action and a collaborative, business-focused mindset.