Applied AI Engineer

This is a fully remote position, open to applicants in Germany.

📋 Description

• Take charge of the architecture for agentic systems.

• Develop the agent harness, which includes orchestration, tool interfaces, sandboxing, permission models, state and memory management, subagents, and tracing.

• Engineer loop processes and context management, focusing on execution rules, budgets, context windows, history compaction, and cross-session memory.

• Create scalable services utilizing Python and TypeScript.

• Design APIs and implement asynchronous processing, including queues, backpressure, rate limits, idempotency, and tenant isolation.

• Construct the data and retrieval framework, encompassing ingestion and normalization through vector and full-text indexing.

• Assess and enhance retrieval quality.

• Perform trajectory evaluations, prompt and tool regression tests, and establish guardrails.

• Manage systems utilizing containerization, CI/CD, and monitoring tools.

• Oversee latency management and model costs.

• Collaborate with coding agents to evaluate models and frameworks.

• Work closely with the Head of AI to decide on the technology stack and on-premises deployment for reasons of data protection or cost efficiency.

• Team up with a small group, engaging directly with the Head of AI.

⛳️ Requirements

• Multiple years of experience in backend or platform engineering.

• Demonstrated experience in operating agentic systems in production environments, beyond mere proof-of-concept projects.

• Familiarity with agent architectures, tool invocation, tool design, MCP, planning and reflection patterns, as well as multi-agent configurations with subagents.

• Proficient in designing execution loops, context compacting, and managing cross-session states.

• Ability to write clean, testable code in Python and TypeScript.

• Experience in developing APIs with FastAPI or Node.js.

• Regular experience working with coding agents.

• Background in systems under load, including concurrency, queues, rate limits, and distributed state management.

• Experience with Docker, CI/CD, and deployments on platforms like AWS, Azure, GCP, or self-hosted infrastructures.

• Knowledge of Kubernetes and local GPU environments is advantageous.

• Experience with ETL and ELT pipelines.

• Proficiency in working with vector databases.

• Understanding of evaluation, tracing, and observability, with a keen awareness of latency and token costs.

• Familiarity with GDPR-compliant architectures, tenant isolation, secrets management, and the boundaries between cloud and on-premises systems.

• Excellent command of German at a C2 level.

• Business-fluent in English.

🏝️ Benefits

• Customized professional development opportunities and conference participation.

• Flexible working hours.

• Fully remote work option within Germany.

• Opportunities for workations.

• Options for sabbaticals.

• Availability of full-time or part-time positions.

• Collaboration with a robust network of experts.

• Regular team building events, company celebrations, and shared activities.

• JobRad bicycle leasing program.

• Modern office locations in Berlin and Augsburg for in-person meetings or team gatherings.

• Access to modern equipment.

• Complimentary drinks and snacks.

• Shared lunch breaks.

• Support for mental and physical well-being through the WorkLifePortal.

• Company pension plan with contributions from the employer.