Senior AI Safety & Fairness Researcher — Remote

MaziCTools; Los Angeles (CA); Hybrid; USD 41,000 - 56,000; Full time

Job summary

Microsoft Exploration seeks a Technical Support Engineer to join a team focused on responsible AI research and mitigation of potential harms from advanced models. You will work with researchers and engineers to develop tooling, implement evaluation methods, and contribute to cross‑functional projects addressing decency and safety in AI.

You will engage with diverse experts and help shape experiments that demonstrate practical, interpretable AI safeguards across real‑world scenarios.

Qualifications

• Bachelor’s degree in measurements, econometrics, software engineering, electrical or computer engineering, or related field.
• 4+ years related experience (e.g., measurements, prescient examination, research).
• Graduate degree (Masters/PhD) with relevant research experience may apply.

Responsibilities

• Identify, estimate, and mitigate potential harms from generative AI systems as part of an interdisciplinary team.
• Develop and use tooling to advance responsible AI research and measurement practices.
• Collaborate with product groups to apply appropriate decency metrics and evaluation methods.
• Present work to internal and external audiences and contribute to research projects.
• Adapt to evolving decency estimation methods and conduct case studies with practical demonstrations.

Skills

Responsible AI

Python

Java

Scala

C#

C/C++

Education

Bachelor's degree in related field

Master's degree in related field

Doctorate in related field

Job description

Microsoft Exploration seeks a Technical Support Engineer to join a team focused on responsible AI research and mitigation of potential harms from advanced models. You will work with researchers and engineers to develop tooling, implement evaluation methods, and contribute to cross‑functional projects addressing decency and safety in AI.

You will engage with diverse experts and help shape experiments that demonstrate practical, interpretable AI safeguards across real‑world scenarios.