Principal Data Scientist

MicrosoftNO

PrincipalFull-time

Posted last monthStill open, checked 1h ago

About the job

As a Principal Data Scientist, you will set the data science direction for some of the world's largest and most advanced services, using data and AI to shape user experiences, improve service quality, and drive measurable impact at global scale. Define and drive the data science strategy and technical roadmap for a product area, identifying the highest-impact opportunities and aligning partners across teams and organisations. Lead the design and application of advanced analytical, AI, and modelling approaches to understand user behaviour at scale, and translate findings into product and platform direction. Own the metrics and evaluation strategy for intelligent services, defining north-star and guardrail metrics, offline and online evaluation frameworks, and quality bars for LLM-powered experiences. Set the technical direction for LLM evaluation, prompt optimisation, and fine-tuning strategies, including scalable evaluation methods (e.g., LLM-as-judge, human labelling, and their calibration). Establish standards for robust, reusable, production-quality code, data pipelines, and experimentation practices, and drive their adoption across teams. Pioneer the use of AI-powered tools and agentic workflows to accelerate analysis, experimentation, and solution development, and scale these practices across the organisation. Influence senior leadership decisions by framing ambiguous problems, presenting clear, evidence-based recommendations, and articulating trade-offs to executives and stakeholders. Lead data strategy for your area, including data acquisition, data quality, governance, privacy, and Responsible AI compliance. Lead cross-functional, multi-team initiatives with local and remote partners, resolving ambiguity and driving alignment to deliver scalable, elegant, and impactful solutions. Mentor and coach data scientists and engineers, contribute to hiring, and foster a culture of scientific rigour, collaboration, and inclusion. Bachelor's, Master's, or Doctorate degree in Data Science, Mathematics, Statistics, Econometrics, Economics, Operations Research, Computer Science, OR a related field AND demonstrated data-science experience (e.g., managing structured and unstructured data, applying statistical techniques, and reporting results); Strong software engineering background covering best practices (code quality, repository hygiene, code reviews, unit testing, design documentation, continuous integration, and deployment), with experience setting engineering standards for data science teams. These requirements include but are not limited to the following specialized security screenings: Customer-facing, project-delivery, professional-services, and/or consulting experience. Proven track record of defining and leading data science and machine learning initiatives from ambiguous problem statements through to measurable production impact, spanning multiple teams or organizations. Deep expertise with large language models (LLMs), including evaluation methodology, prompt engineering, fine-tuning, and grounding/retrieval-augmented generation for real-world applications at scale. Expertise in experimentation design (e.g., A/B testing, causal inference) and in defining metrics that capture end-user impact. Extensive experience with distributed big-data computing systems (e.g., Spark, Hadoop, Hive, Azure ML). Expert-level proficiency with ML tools to build models and analyses data (e.g., Python, R, scikit-learn, TensorFlow, PyTorch, ML.NET). Demonstrated ability to influence technical direction without direct authority and to communicate complex findings to executive audiences. Experience mentoring and growing senior data scientists. Publications, patents, or recognized contributions to the data science or machine learning community.