Functieomschrijving

We're on the lookout for a Mid-Level AI Systems Engineer who bridges AI engineering and MLOps. Someone who loves getting hands-on with LLMs, agents, and RAG pipelines, and who cares just as much about what happens once a model reaches production as about building it in the first place. In this role, you'll design, build, deploy, and operate production-grade AI systems, with a strong focus on generative AI, LLM-based applications, and reliable machine learning operations. In practice your time is split roughly between 60-70% AI Engineering and 30-40% MLOps.

In this role you sit at the intersection of engineering and operations, working across teams and disciplines. You'll report to the Head of ML at Sagacify and collaborate with the wider Sagacify and Craftzing delivery organisation. A role that can naturally grow towards a Team Lead position over time.

What you'll do :

AI Engineering (60-70%)

• Orchestrate AI system components: LLMs, vector databases, APIs, orchestration layers, and user interfaces
• Develop autonomous agents and conversational systems that can plan actions and interact with external tools or APIs
• Build and optimise RAG pipelines connecting enterprise data sources to LLMs for grounded, contextualised, reliable answers
• Evaluate system quality through generative AI metrics, coherence tests, and production monitoring (latency, API costs, bias)
• Deploy and scale solutions with strong attention to latency, security, reliability and cost efficiency
• Keep an eye on the ecosystem for new models, frameworks and techniques, including open-source tools such as LangChain, LangGraph, Langfuse, etc.
• Perform prompt and context engineering to improve output quality, reduce hallucinations, and manage conversational state effectively

MLOps (30-40%)

• Build and maintain automation for model deployment, including CI/CD pipelines and automated testing
• Continuously monitor model performance in production, including drift detection and quality metric tracking
• Manage updates of libraries, models, and related dependencies in production environments
• Ensure versioning, reproducibility and safe rollout of models and AI services
• Collaborate closely with ML engineers, developers, DevOps and infrastructure teams for smooth delivery
• Stay current with the latest MLOps practices, tools and platform components

You stay in the code, you stay curious about what happens after deployment, and you keep learning as you go.