Data Scientist / Applied Scientist (Mid and Senior Levels)

Share job

We are hiring multiple Data Scientists and Applied Scientists across LATAM to join the Microsoft 365 team. These are remote positions, allowing you to work from the comfort of your home!

As part of our team, you will help shape one of Microsoft’s fastest-growing cloud services. Your work will directly impact Copilot, enabling personalized, context-aware experiences that empower millions of users across Microsoft 365.

Do you want to collaborate closely with software engineers, product managers, and researchers to design, implement, and evaluate models and algorithms that continuously adapt and improve based on user interactions and feedback? Do you want to leverage advanced machine learning and information retrieval to predict and optimize memory content selection, representation, and lifecycle management.

You will work at the forefront of applied AI and machine learning, shaping the future of intelligent productivity tools. You will contribute to a culture of innovation, continuous learning, and experimentation.

About the team

The Substrate Core Substrate team powers the infrastructure that underpins Microsoft 365’s most critical services and Copilot. We are a high-impact, forward-looking team focused on building intelligent, scalable, and cost-efficient platforms that enable Microsoft to deliver world-class productivity experiences to billions of users. Our mission is to drive operational excellence and innovation across the M365 fleet by leveraging AI, automation, and deep platform integration.

Your core mission will be to deliver exciting innovation in Microsoft Copilot. Microsoft Copilot is revolutionizing how people work and have created an unprecedented opportunity to advance the state-of-the-art in a way that benefits millions of people.

More about your Responsibilities

We are seeking candidates with research skills and the desire to pursue the cutting edge in model development that pushes technological boundaries. We are looking for candidates with interest and experience in language model training, large language model evaluation, and quality assessment. In particular we would like a candidate that has a passion for understanding what makes an AI solution work and how to measure its performance.

We are looking for candidates who are creative, self-driven, curious, people-oriented and comfortable defining a path through ambiguity towards high-level goals. The successful applicants will collaborate closely with people across Microsoft’s product teams and Microsoft Research to see their ideas realized in products and services that are used worldwide, as well as to contribute back to the scientific community and beyond (e.g. via presentations, publications, media interviews, and similar). Come to define, design, and deliver what enterprise cloud productivity means.

As a Data Scientist / Applied Scientist, you will help transform how Copilot features are evaluated and improved. Your team will deliver end-to-end experimentation, evaluation, and insights to Copilot engineers, PMs, and fellow scientists. You’ll work on the data generation platform that creates algorithms and ML pipelines for simulating user actions and datasets that reflect real-world usage and user preferences and develop metrics that go beyond accuracy – capturing nuance, intent, and satisfaction. You’ll work on scalable pipelines that support offline evaluations for Copilot Search, BizChat, Connectors, and Agents (DAs). This opportunity will allow you to dive deep into Copilot technologies, shape how AI quality is measured at scale, build critical skills in the AI-era, and rapidly grow your career.

By applying to this position, you will be considered for multiple similar positions within our organization for an invitation-only virtual interview day. Position specifics, including hiring team, location, and position details, will be determined following the interview process.

Main qualifications

• Bachelor’s degree in Statistics, Econometrics, Computer Science, Electrical/Computer Engineering, or related field.
• 4+ years of experience in predictive analytics, statistics, or research.
• Experience with synthetic data generation and data management for evaluation/training.
• At least one year of experience publishing patents or peer-reviewed papers.
• Deep motivation for user-centric AI and interest in human cognition, memory, and AI.
• Global collaboration is integral to our work at Microsoft, and proficiency in English is essential for effective communication in this role.

Preferred qualifications

• Master’s degree in Statistics, Econometrics, Computer Science, Electrical/Computer Engineering, or related field.
• Experience with large-scale embedding models and transformer architectures.
• Familiarity with reinforcement learning and distributed computing platforms (e.g., Heron, AML, Euclid).
• Proficient analytical skills and experience with telemetry and performance metrics.
• DevOps experience and cloud services knowledge (Azure preferred).
• Agile development experience and a structured approach to software design.

Interpersonal skills

• Growth mindset with resilience and a passion for learning.
• Commitment to diversity, inclusion, and authentic collaboration.
• Effective communication skills with the ability to engage diverse audiences.
• Ownership mindset with the ability to deliver results and manage expectations.
• Self-motivated and collaborative across teams and organizations.

• Develop and evaluate ML models using prepared datasets, customer feedback, and novel training/fine-tuning algorithms for language models.
• Write production-quality code, apply debugging best practices, and stay current with industry trends.
• Drive customer-centric solutions by aligning with business goals and managing stakeholder expectations.
• Collaborate cross-functionally to define success metrics and improve AI quality at scale.
• Lead research projects that yield new algorithms, tools, or insights solving open problems.
• Analyze evaluation outputs to identify gaps in coverage, quality, and usability.
• Design experiments, define metrics, and develop ML pipelines for encoder-decoder and cross-encoder models, semantic search, and user intent understanding.
• Work with large-scale data while championing privacy and compliance.
• Engage with customers and internal teams to identify pain points and drive impactful improvements.
• Stay engaged with the latest research and foster a strong applied science culture. Publishing is encouraged.