10 Anthropic AI safety researchers
Inside Anthropic, these AI safety researchers name alignment stress-testing, interpretability and scalable oversight as their own research areas; all ten are in current roles, seven with Member of Technical Staff titles and two as research scientists. They suit teams designing model evaluations, red-teaming a model before release, or deciding how much to rely on interpretability methods.
Professionals to explore
Ajay Garg
LinkedInExperience: Research Scientist, Interpretability · Anthropic
$2,019 per 30-minute call
Research Scientist, Interpretability at Anthropic since 2024, Ajay Garg was a Staff machine learning Engineer at Stealth (2023-2024), a Senior Machine Learning Engineer at Amazon (2022-2023) and a Machine Learning Engineer - NLP at Apple (2018-2021). The interpretability title and NLP engineering background fit questions on applying interpretability tools to language models.
Is this you? Claim your profile and get paid for calls.
Alex Rodrigues
LinkedInExperience: AI Alignment Research · Anthropic
$1,346 per 30-minute call
Alex Rodrigues has worked in AI Alignment Research at Anthropic since 2024, describing a focus on AI aligned with human values. Alex was CEO and Co-Founder of Embark Trucks (2015-2023) and a Thiel Fellow (2016), which adds the view of someone who has been a company's CEO and co-founder.
Is this you? Claim your profile and get paid for calls.
Amanda Askell
LinkedInExperience: Member Of Technical Staff · Anthropic
$6,058 per 30-minute call
Amanda Askell, a Member Of Technical Staff at Anthropic since 2021, works on aligning and evaluating AI systems with a focus on large language models. Amanda was Research Scientist (Policy) at OpenAI (2018-2021) and has a background in formal ethics and decision theory, so questions about model values, evaluation and AI policy fit well.
Is this you? Claim your profile and get paid for calls.
Daniel Ziegler
LinkedInExperience: Member of Technical Staff · Anthropic
$5,385 per 30-minute call
On Anthropic's Alignment Stress-Testing work, Daniel Ziegler has been a Member of Technical Staff since 2024. Daniel also held Member of Technical Staff titles at METR (2023-2024), Redwood Research (2021-2023) and OpenAI (2018-2021), giving a view of research practice across four organizations.
Is this you? Claim your profile and get paid for calls.
Euan Ong
LinkedInExperience: Member of Technical Staff (Alignment Science, Interpretability) · Anthropic
$1,683 per 30-minute call
Euan Ong holds the title Member of Technical Staff (Alignment Science, Interpretability) at Anthropic (since 2024). Euan was a Research Assistant at the University of Cambridge (2023-2024) and a Research Intern at the Center for Human-Compatible AI (2023), which suits questions connecting interpretability research to alignment work.
Is this you? Claim your profile and get paid for calls.
Evan Hubinger
LinkedInExperience: Member of Technical Staff, Manager · Anthropic
$2,693 per 30-minute call
Evan Hubinger has been Member of Technical Staff, Manager at Anthropic since 2023, described as the Alignment Science org lead focused on red-teaming Anthropic's alignment work, and leads the Alignment Stress-Testing team. Evan was also a Member of Technical Staff (Intern) at OpenAI in 2019, and is the most direct contact here for stress-testing alignment techniques.
Is this you? Claim your profile and get paid for calls.
Harrison Lee
LinkedInExperience: Member of Technical Staff · Anthropic
$1,683 per 30-minute call
Based in the United Kingdom, Harrison Lee describes being an alignment researcher and has been a Member of Technical Staff at Anthropic since 2026. Harrison was a Staff Research Scientist at Google DeepMind (2020-2026) and a Machine Learning Engineer at Facebook (2017-2020), a fit for comparing alignment research practice at Google DeepMind and Anthropic.
Is this you? Claim your profile and get paid for calls.
Joshua Batson
LinkedInExperience: Research Scientist · Anthropic
$2,693 per 30-minute call
Joshua Batson is a Research Scientist at Anthropic (since 2023), studying how neural nets work inside and what that means for performance and safety. Joshua was a Senior Research Scientist at Biohub (2017-2021) and Chief Data Scientist at PHC Global (2021), making interpretability and its use in safety decisions the right topics.
Is this you? Claim your profile and get paid for calls.
Liang Qiu
LinkedInExperience: Member of Technical Staff · Anthropic
$1,683 per 30-minute call
Scalable oversight is Liang Qiu's stated focus as a Member of Technical Staff at Anthropic since 2026. Liang was a Senior Applied Scientist (2024-2025) and Applied Scientist (2022-2024) at Amazon and holds a Ph.D. from UCLA, so questions about supervising models on tasks humans struggle to check fit best here.
Is this you? Claim your profile and get paid for calls.
Sara Price
LinkedInExperience: Member of Technical Staff · Anthropic
$1,683 per 30-minute call
Sara Price has been a Member of Technical Staff at Anthropic since 2025 and was an Independent AI Alignment Researcher with Anthropic (2024-2025), describing the work as AI safety and alignment research. Sara was also a Research Scholar at MATS Research (2023-2024), useful for questions on how independent research feeds into a lab's safety work.
Is this you? Claim your profile and get paid for calls.
Choose the right perspective
For red-teaming and stress-testing alignment techniques, Evan Hubinger and Daniel Ziegler are the closest fit. Joshua Batson, Ajay Garg and Euan Ong cover interpretability, Liang Qiu covers scalable oversight, and Amanda Askell, Harrison Lee, Sara Price and Alex Rodrigues fit broader alignment and evaluation questions.
Questions to take into the conversation
- 01How do you design a stress test that would catch a model behaving differently when it thinks it is being evaluated?
- 02Which interpretability findings are mature enough to inform a decision about deploying a model?
- 03How does scalable oversight work when the task is too hard for human reviewers to check directly?
About this directory
This is a professional research starting point based on business profile data retrieved on . Titles and companies reflect that source snapshot; roles marked former had ended by then. Check the linked profiles for current details. Inclusion does not imply Instant Expert membership or availability.