AI Engineer – LLM Apps, RAG & Agentic Systems, German-speaking

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

📋 Description

• You will create the intelligence that powers ArcGEN.

• You are accountable for our LLM pipelines, agent architectures, and evaluation frameworks, overseeing everything from prototype notebooks to production-ready pipelines with quantifiable quality.

• Design and manage RAG pipelines, including embeddings, vector stores, and re-ranking.

• Architect agentic systems, focusing on tool-calling, MCP, and sub-agent orchestration.

• Develop and sustain LLM evaluation frameworks, including metrics, benchmarks, A/B testing, and quality gates.

• Engage in prompt engineering and systematic prompt optimization.

• Create ML models tailored for marketing applications, such as attribution, targeting, forecasting, and anomaly detection.

• Transform notebook explorations into production-quality pipelines.

• Collaborate closely with our full-stack engineers for seamless platform integration.

⛳️ Requirements

• Practical experience with modern LLM production workflows, including RAG, embeddings, and vector databases like Pinecone, Weaviate, and pgvector.

• Strong Python proficiency, particularly with pandas, NumPy, scikit-learn, and either PyTorch or TensorFlow.

• Familiarity with the complete ML lifecycle, from data preprocessing to training, evaluation, deployment, and monitoring.

• Experience in systematic model evaluation, including metrics, benchmarks, and evaluation frameworks.

• Proficient in German at C2 level and English at C1 level.

• A builder mindset: you deliver production systems rather than just notebooks.

• Experience in developing custom agents, sub-agents, or specialized coding workflows is a plus.

• Familiarity with MCP (Model Context Protocol) and tool-calling architectures is an advantage.

• Knowledge of agent frameworks such as Mastra, LangGraph, or Vercel AI SDK is a plus.

• Experience with fine-tuning, LoRA adapters, or distillation is advantageous.

• Familiarity with MLOps tools like MLflow, Weights & Biases, or DVC is a plus.

• Background in marketing analytics, attribution models, or performance marketing data is an advantage.

• Contributions to Kaggle, open-source projects, or publications in ML/AI are a plus.

🏝️ Benefits

• Competitive salary with yearly adjustments.

• Fully funded access to Claude Max, Codex, GitHub Copilot, OpenAI and Anthropic APIs, as well as all relevant LLM tools.

• A compute budget dedicated to experiments, model training, and ML workloads.

• A MacBook Pro of your choice.

• Remote-first work environment with flexible hours.

• Quarterly team meetups in Berlin or Dubai (DIFC AI Campus, optional).

• Opportunity to directly influence product architecture at an early stage.

• A steep learning curve in agentic AI, LLM orchestration, and AI-native engineering.