Senior AI Safety Researcher

Written by the T&S Network from Google's posting.

Builds safety protections for foundation models and their downstream products, including Gemini, Veo, Agents, Robotics, enterprise Vertex AI and on-device applications. The AI Safety Protections team within Trust and Safety works on frontier risks from more capable models that could cause more severe harm if misused or misaligned, and sits between research, model development and policy.

• AI misuse and jailbreaks

What you'd do

• Build scalable safety systems for AI products using advanced machine learning methods.
• Apply statistical and data science methods to examine existing protection measures and find their shortcomings.
• Turn threat models into evaluations, mitigations and launch decisions.
• Build data stories for stakeholders, including senior leadership, to drive outcomes.
• Mentor and guide junior team members to support their growth.
• Work with sensitive content and situations, including graphic, controversial or upsetting material.

What it asks for

• Five years analyzing data, spotting trends and drawing insights from both quantitative and qualitative sources.
• Five years of experience in data analysis or data science.
• Experience with programming languages such as C/C++, Python or Java, and SQL.
• Master's degree or PhD in a quantitative field such as computer science, statistics or mathematics (preferred).
• Experience in trust, safety, abuse or fraud, especially web security, harmful content moderation or threat analysis (preferred).
• Experience building and shipping machine learning models at scale (preferred).

Worth knowing

• Exposure to graphic, controversial or upsetting content.

Our best summary of the posting, and it can be wrong. The employer's posting is what counts: read it in full on Googlenorth_east before you apply. If we got something wrong, report this listing.