Research Fellow, AI Interpretability

Goodfire; United States; On-site; USD 22,000 - 56,000; Full time

Job summary

Effectief Altruisme is seeking a three-month AI interpretability research fellow to work on a predetermined project under mentorship from leading researchers. You will execute the project, iterate based on feedback, and deliver a final report.

The role requires a strong technical background in CS or ML, Python proficiency, and experience with ML frameworks. This fellowship offers hands-on research with practical impact in model interpretability.

Qualifications

• Strong technical background in computer science or machine learning.
• Proficiency in Python and ML frameworks.
• Prior research experience is required.

Responsibilities

• Execute an assigned interpretability research project and produce a research report by completion.
• Implement mentor feedback while maintaining independent execution and fast iteration.
• Contribute to the understanding of how large AI models behave and why they act as they do.

Skills

Python

ML frameworks

research experience

CS/ML background

Job description

A three-month AI interpretability research fellowship executing focused projects under mentorship from leading researchers.

• Execute an assigned interpretability research project and produce a research report by program completion.
• Implement feedback from mentors while maintaining independent execution and fast iteration.
• Requires strong technical background in computer science or machine learning with prior research experience.
• Requires Python proficiency and ML frameworks; foundation…

Grote AI-modellen worden getraind door te proberen en bij te sturen, zonder dat iemand goed begrijpt waarom ze doen wat ze doen. Als fellow werk je drie maanden aan een vooraf bepaald onderzoeksproject dat dit zichtbaar moet maken. De hefboom zit in je eindverslag: als een methode of bevinding werkt, kunnen andere onderzoekers die overnemen om modellen beter te doorzien en veiliger te maken.