Public summary
Join a fast-growing Berlin-based startup building an AI-native platform for accounting, payroll, and tax for European SMEs. The role focuses on automating manual workflows across teams using AI, working directly with the founders. The company employs advanced SDLC2 methodology and emphasizes fast prototyping and robust agent development. English fluency is required; German is a plus but not mandatory. The company offers an above-market salary, flexible and autonomous working conditions, modern office in Berlin Mitte, comprehensive benefits, and a dynamic, diverse team environment.
Location and work setup
- Location
- Berlin
- Remote status
- On-site
- German requirement signal
- No German Required Detected
- Detected job language
- English
Responsibilities
Develop and deploy one automation agent weekly focusing on high-friction workflows to save manual effort. Define and maintain organizational sources of truth including SOPs, knowledge bases, and code access for AI agents. Manage context drift and documentation accuracy by deploying agents that monitor and flag discrepancies. Build and maintain integration servers and tools for accessing internal systems. Develop and continuously evaluate agents using real-world failure cases, monitor their performance, and iterate accordingly.
Qualifications
Minimum 4 years of experience shipping production software with full-stack TypeScript or Python expertise. At least 1 year experience developing and shipping LLM applications with formal evaluation processes. Demonstrated experience building MCP servers or equivalent. Deep understanding of AI models, frameworks, and best practices for agent-based automation. High contextual awareness and ability to engage with multiple teams to translate workflows into AI agents. Strong decision-making skills and comfortable working autonomously. English fluency is required; German language skills are a plus but not mandatory.