TL;DR

Applied Scientist 2 (AI): Developing a next-generation grounding service to provide factual and trustworthy responses for AI applications like chat assistants and copilots, focusing on retrieval, reasoning, and real-time data integration. Focus on inventing retrieval and attribution methods, defining factuality metrics, and shipping production models that scale to billions of queries.

Location: Suzhou, China

Company

Microsoft’s mission is to empower every person and every organization on the planet to achieve more.

What you will do

Own the science roadmap for grounding, including retrieval, re-ranking, attribution, and reasoning.
Build citation and provenance systems to reduce hallucinations and increase user trust.
Advance tool-augmented grounding through schema-aware retrieval and real-time connectors.
Partner with product and business leaders to influence strategic decisions and drive business impact.
Mentor applied scientists and data scientists, establishing best practices in experimentation and error analysis.
Contribute to ethics and privacy policies, identifying bias in product development, and propose mitigation strategies.

Requirements

Bachelor’s Degree in Computer Science, Electrical or Computer Engineering, or related field AND 2+ years related experience OR Master’s Degree AND 1+ year of experience OR Doctorate.
Minimum of 2 years of hands-on experience designing and building search, retrieval, or ranking systems.
Proven track record of shipping LLM-powered or Retrieval-Augmented Generation (RAG) systems into production environments.
Solid coding skills and solid foundation in machine learning.
Demonstrated ability to lead through ambiguity and deliver measurable impact in cross-functional settings.

Nice to have

Bachelor’s Degree AND 5+ years related experience OR Master’s Degree AND 3+ years related experience OR Doctorate AND 1+ year(s) related experience.
Minimum of 4 years of hands-on experience designing and building search, retrieval, or ranking systems.
Demonstrated expertise in information retrieval with publications in top-tier conferences or journals.
Hands-on experience in large language model (LLM) development, including pretraining, supervised fine-tuning (SFT), and reinforcement learning (RL).
Proven track record in optimizing LLM inference, or active contributions to open-source frameworks.

Culture & Benefits

Growth mindset and inclusive collaboration are valued.
Innovate to empower others and collaborate to realize shared goals.
Build on values of respect, integrity, and accountability to create a culture of inclusion where everyone can thrive.