Senior Data Scientist, M365 Copilot, Agents & Platform

What you'll do

• Write Python, SQL, and KQL code to prepare and analyze large-scale customer feedback and multimodal logs.
• Build representative evaluation datasets for diverse scenarios including information retrieval, coding, and technical support.
• Design evaluation coverage for text and voice inputs across single- and multi-turn conversations.
• Own experiment design and impact measurement for optimization hypotheses, including success metrics and guardrails.
• Conduct side-by-side evaluations to compare LLM-based and human results while assessing statistical significance.
• Develop decision-relevant metrics and dashboards to translate complex analyses into actionable product recommendations.
• Present analytical findings and limitations to technical and executive audiences to guide key product initiatives.
• Provide mentorship through analytical code reviews and guidance on study design and data science best practices.

What we're looking for

• Must have a Doctorate and 1+ years of data science experience, a Master's and 3+ years, or a Bachelor's and 5+ years of experience.
• Must have equivalent experience to the degree and years of experience requirements.
• Must be able to pass the Microsoft Cloud background check.
• Must be able to write efficient code in Python, SQL, KQL, or similar languages to analyze large-scale datasets.
• Must be able to design experiments, build evaluation datasets, and translate results into actionable product recommendations.
• Must be able to communicate technical details clearly to both technical and executive audiences.
• Must have 4+ years of experience using Python, SQL, KQL, R, or similar languages to analyze large-scale structured and unstructured datasets (preferred).
• Must have experience driving cross-organizational communication and end-to-end analytical problem solving in complex domains (preferred).