Key Skills

Job Description

About This Vacancy As a Data Scientist, you will develop intelligent solutions based on operational, process and asset data. Depending on the project, your work may include: Developing machine learning and predictive models for process optimization, anomaly detection and predictive maintenance. Building Generative AI and Agentic AI applications that retrieve information, reason across multiple steps and interact with tools, APIs and data sources. Designing RAG solutions using structured data, documents, events, embeddings, vector/hybrid search and knowledge graphs. Developing multi-agent and tool-using workflows, including decision logic, memory orchestration and human-in-the-loop controls. Applying process mining, anomaly investigation and root-cause analysis to operational challenges. Exploring new use cases and assessing data availability, data quality and potential business value. Translating analytical results into actionable insights and recommendations together with domain experts and end users. Building reusable AI services and APIs that integrate with operational data and digital products. Productizing prototypes using solid software-engineering practices, including automated testing, CI/CD and observability. Supporting the deployment and integration of machine learning and AI solutions into production environments. Exploring new developments in Generative AI, machine learning, process mining and knowledge graphs, and translating promising technologies into practical applications.

Core Responsibilities

Develop and productize machine learning, Generative AI, and agentic solutions for operational optimization, anomaly investigation, and predictive maintenance. Integrate reusable AI services with operational data and digital products, support production deployment, and translate analytical findings into actionable recommendations with domain experts and end users.

Requirements

The role calls for expertise in machine learning, Generative AI, RAG, process mining, and knowledge graphs, along with the ability to assess data quality and business value. It also involves building production-ready software using practices such as automated testing, CI/CD, and observability, and collaborating with domain experts and end users.