Software Engineer, Back-End (LLM Agents)

Superhuman is looking for a back-end software engineer to build LLM-powered, agentic AI features end-to-end. The role involves designing reliable, fast, and scalable services, protocols, and data flows, taking ownership of ambiguous problems, and shipping in tight iterations. The ideal candidate is back-end-strong yet stack-flexible, with hands-on experience integrating LLM/agentic features and strong fundamentals in Java, Go, or Python. This hybrid role is based in Berlin and includes relocation support.

• Build LLM-powered and agentic AI features end-to-end, from idea to production.
• Design services, protocols, and data flows that make AI features reliable, fast, and scalable.
• Build and support production services with high call rates, targeting high availability.
• Take on ambiguous problems—shaping scope, estimating work, and making quality, cost, and latency trade-offs.
• Move across the stack when needed, from client platforms to shared protocols to infrastructure.
• Learn infrastructure as code and ship in tight iterations.
• Mentor new hires and help raise the engineering bar across the team.

• 5+ years of relevant experience in back-end development.
• Back-end-strong but stack-agnostic, comfortable working across the stack.
• Hands-on experience building or integrating LLM-powered or agentic features.