Staff Applied Research Scientist

Responsibilities

• Lead research projects that apply formal methods, program analysis, automated reasoning, and AI-driven techniques (including code generation and modeling) to real problems in our cloud data platform.
• Translate research ideas into prototypes, then into shipped capabilities that move concrete business metrics — quality, velocity, reliability, and operational performance at scale.
• Partner closely with engineering leaders, product managers, and key customers to identify high-leverage opportunities and turn them into deliverables.
• Influence the engineering and product roadmap; advise leaders on which research directions are pragmatic and which are not.
• Train and uplevel engineering teams on new methods, and scale those methods across the organization.
• Maintain expertise at the frontier of the field through publications, conference participation, open-source contributions, and patent filings.

Requirements

• PhD (or equivalent research experience) in Computer Science or a closely related field.
• Depth across the areas this role sits at the intersection of:
  • Formal methods — e.g., model checking, theorem proving, SAT/SMT, program verification, type systems, or program analysis.
  • Distributed systems — designing, reasoning about, or verifying large-scale concurrent and distributed systems.
  • Software engineering — strong fundamentals; able to go from a research idea to production-quality code in collaboration with engineering teams.
  • AI / ML — practical experience applying modern ML, including LLMs, to systems problems such as code generation, synthesis, or automated reasoning.
• 8+ years applying theoretical computer science to large-scale software systems — ideally cloud data platforms, distributed systems, or developer infrastructure.
• Demonstrated ability to drive company-level initiatives in partnership with engineering and product leadership.
• Track record of technical contribution to the field — publications, open-source work, patents, or comparable evidence of impact.
• Comfortable in a fast-paced, ambiguous environment where impact is measured by what ships.