Senior Machine Learning Engineer, MLOps

Job ID: 3497

 

Are you passionate about shaping the future of insurance technology? We are looking for a Senior ML Engineer who can turn machine learning from promising prototypes into reliable production services and help build long-term technical quality in a regulated business environment.

 

Welcome to Nordea Life & Pension Finland. We are a solvent Finnish life insurance company and part of the Nordea Group, but we are not Nordea Bank’s everyday banking business. Our focus is on life insurance, savings and pension-related products, and while Nordea Bank acts as an important distribution channel and agent for several of our products, Nordea Life & Pension has its own business domain, products and dedicated IT organisation.

 

About this opportunity

 

Nordea Life & Pension Finland combines the strengths of a specialised insurance company with the backing of a large financial group. We are much smaller than Nordea Bank, which means closer collaboration, shorter decision paths and strong ownership in day-to-day work. At the same time, because we are part of the Nordea Group, the role also requires the ability to work across shared platforms, common processes and a larger enterprise environment. We are a small company recognized repeatedly as Finland's best financial sector workplace by Great Place to Work and winner of the Finnish Quality Award. Quality is not a project; it is part of how we think. We do things properly, improve continuously, and keep structures light so people can focus on meaningful work.

 

Now we focus on building our AI capability over maintaining a legacy. You'll join alongside data scientists, developers, and MLOps. Technical ownership, including architecture, standards and engineering direction, belongs to the senior engineers we're bringing in. You'll have real influence on how this team operates as it grows. Your focus is the layer between "the model works in development" and "the model works reliably for real customers." Model serving architecture, deployment pipelines, ML-specific monitoring, cost optimization, and the automation that lets the team deploy confidently and roll back safely. You build it, you own it in production, but with platform support from the wider Nordea Group. You won't be setting up Kubernetes clusters, but you'll be responsible for your solutions running reliably. We're building this operational muscle together and value your input on how to structure it well.

 

What you’ll be doing:

• Own the production lifecycle of ML/AI models: serving, monitoring, scaling and reliability
• Design and operate model serving infrastructure (vLLM or equivalent) across hybrid environments
• Implement ML-specific monitoring: prediction quality, drift detection, latency, throughput, and cost per inference
• Manage GPU infrastructure: provisioning, scheduling, utilization optimization
• Ensure security, compliance, and reliability for AI workloads in a regulated environment

 

Our AI workloads include LLM serving on vLLM with GPU infrastructure, agent-based systems, and we use SageMaker for parts of our ML pipeline. The broader stack is Python, AWS, Docker, Kubernetes, Terraform, Jenkins, and Git. We prefer open-source and operate in both cloud and on-premises environments.

 

This role is located in our Helsinki (Kaisaniemi). Ideal starting month is as soon as possible.

 

Who you are

 

You understand ML systems deeply: not only how to ship a container, but why machine learning behaves differently from traditional software in production. You know how prediction quality degrades, how data and environment changes affect outcomes, and how to build systems that detect issues before customers notice them.

 

We are looking for someone who has:

• Experience running ML/AI models in production
• Strong Python skills for production systems, not only notebooks or prototypes
• Strong understanding of why ML systems fail in practice: drift, data quality, silent degradation
• Hands-on experience with Docker and cloud infrastructure, ideally AWS, and with bringing ML systems from prototype to reliable production use
• Familiarity with model serving concepts: latency, reliability, scaling
• Working proficiency in Finnish and English
• Master’s or PhD in a related field