Senior Platform Engineer – Agentic AI

paiqo GmbH; Benhausen; Hybrid; EUR 70.000 - 90.000; Vollzeit

Zusammenfassung

Ein innovatives Unternehmen im Bereich AI sucht einen erfahrenen Senior Agentic AI Engineer. Der Kandidat sollte über mindestens 5 Jahre Erfahrung in der AI-Entwicklung verfügen und Kenntnisse in Python sowie Erfahrungen mit RAG-Architekturen mitbringen. Flexible Arbeitszeiten und ein hybrides Arbeitsmodell werden angeboten.

Leistungen

Flexible Arbeitszeiten

Budget für AI-Training und Zertifizierungen

Diversity und Inklusion

Qualifikationen

• Mindestens 5 Jahre Erfahrung in AI/ML Engineering oder Softwarearchitektur.
• Gute Kenntnisse in Python und generativen Systemen.
• Fähigkeit, komplexe Themen verständlich zu kommunizieren.
• Fließend Deutsch und sehr gutes Englisch.

Aufgaben

• Design und Implementierung von fortgeschrittenen Agentenarchitekturen.
• Nutzung von Azure AI Foundry zur Orchestrierung komplexer Workflows.
• Entwicklung von Monitoring-Frameworks für Agenten.
• Mentoring von Junior-Teammitgliedern.

Kenntnisse

Python

RAG-Architekturen

Multi-Agenten-Frameworks

DevOps/MLOps

Containerisierung (Docker, Kubernetes)

Kommunikationsfähigkeit

Tools

Azure AI Foundry

Terraform

Bicep

Jobbeschreibung

Want to take AI from chatbots to autonomous agents?

We are pioneers in Agentic AI: our platforms combine powerful LLMs with multi-agent orchestration and responsible AI practices. As a Senior Agentic AI Engineer, you will implement complex architectures – from ReAct strategies and GraphRAG to tool use and reflection. You ensure that agents act reliably, securely, and transparently, thereby driving our customers' success.

Tasks & Responsibilities

• Design and implementation of advanced agent architectures: multi-agent systems with autonomous planning and reflection loops, GraphRAG, and tool orchestration
• Use of Azure AI Foundry, Microsoft Agent Framework, and open-source libraries (LangChain, CrewAI) to orchestrate complex workflows; integration of enterprise data via RAG
• Development of monitoring and observability frameworks for agents, including telemetry, logging, and performance evaluations.
• Developing and enforcing responsible AI principles: data protection, ethics, fairness, and security
• Mentoring junior team members; technical leadership and reviews; participation in architecture roadmaps
• Working closely with customers and management to identify use cases and evaluate business benefits

What we offer you

• Flexible working: Trust-based working hours and remote/hybrid model, provided you are based in Germany
• State-of-the-art tech stack: Access to Azure AI Foundry, Copilot Studio, LangChain & LangGraph and our data lakehouse (Delta Lake, Fabric)
• Investment in your future: Budget for AI training, certifications, and conference attendance; participation in research projects
• Diversity & Purpose: We actively promote diversity, equality, and inclusion; our projects support sustainable and socially relevant goals
• Leadership influence: You will shape technical strategies, establish governance processes, and drive innovation in a rapidly growing team

As a dynamic company, we encourage independent work, continuous learning, and a culture where everyone can contribute ideas. If you want to be at the forefront of Agentic AI development while experiencing a work environment that combines transparency, diversity, and well-being, we look forward to receiving your application!

Qualifications

• At least 5 years of experience in AI/ML engineering, software architecture, or data-driven product development; extensive practice with generative and agentic systems
• Very good knowledge of Python; experience with RAG architectures, vector databases, GraphRAG, prompt engineering, and multi-agent frameworks
• Familiar with DevOps/MLOps methods and infrastructure-as-code (Terraform, Bicep) as well as containerization (Docker, Kubernetes)
• Ability to communicate complex issues in a way that is understandable to non-technical stakeholders
• Fluent German and very good English (spoken and written)