Member of Technical Staff

NOTE: If you previously applied to one of our Research Engineer/Scientist, Machine Learning Research Engineer/Scientist, or Research Stream Lead roles,
you do not need to apply again
. We are merging all inbound applications for researcher roles into this one.

We are a nonprofit research organization that develops scientific methods to assess AI capabilities, risks and mitigations, with a specific focus on threats related to autonomy, AI R&D automation, and alignment.

We believe it is robustly good for civilization to have a clearer understanding of what dangers AI systems pose, and we are extremely excited to find ambitious, excellent people to join our team and tackle one of the most important challenges of our time.

We evaluate candidates primarily through work tests. We usually do an in-person trial as well but can be flexible about this.

METR currently has 3 primary research streams:

• Capabilities: Accurately measuring frontier model performance on threat-relevant tasks (autonomy, AI R&D automation, etc.) and predicting future capabilities. We develop and maintain benchmarks, diverse evidence-gathering methods, and metrics to track capability trends and anticipate the thresholds that matter most for safety.
• Monitorability: Understanding how well frontier models can take subversive or unwanted actions despite various monitoring or control protocols. We build the research infrastructure – novel metrics, control evaluations, elicitation methods – needed to improve the world's understanding of how effectively current and future models can circumvent oversight.
• Alignment/Propensity: Determining whether or not a model that is capable of causing catastrophic harm (in its actual deployment setting) would be likely to actually do so in a given high-stakes deployment setting. We aim to develop the science of propensity evaluations and examine when we might expect high-stakes catastrophic misalignment.

The Capabilities and Monitorability streams are both hiring Research ICs (individual contributors), while the Alignment/Propensity stream is hiring for a Research Stream Lead, followed by Research ICs down the line. The stream you end up joining will be based on a combination of working fit and interest.

For our Research IC roles, we are looking for a combination of skills across “research science”, “research execution” and software engineering. You may not have all of these skills (for example, we don’t expect software engineering to be a large part of the role for narrowly focused researchers). For the Research Stream Lead role, we are additionally looking for research management skills.

• You have strong knowledge of relevant literature and general research good practice.
• You have good understanding of how particular projects fit into METR's overall mission – you are thinking about things like “how will this generalize to future models”, or “how does this relate to alignment evals”.
• You reliably notice important but subtle methodological limitations.
• You are undaunted by open-ended mandates – you can take a confusing or ill-posed question and produce insightful and helpful frameworks / proposals / results.
• You can write great papers.

• You are an experienced executor/contributor; you are familiar with patterns of successful and unsuccessful execution in frontier ML research. You are undaunted by “I’ve never done this before” or even “no-one has done this before”.
• You are creative, ambitious and entrepreneurial. You work fast and are highly responsive and available. You can juggle many balls when it is useful.

• You balance rapid prototyping with the creation of maintainable, scalable systems and make sound technical decisions.
• You lead large projects from ideation to delivery, balancing innovative ML solutions with reliable, high-quality code.
• You set high standards for system architecture, code quality, and maintainability, influencing broad software practices across the organization.
• We're seeking a researcher to help us better understand AI capabilities. Previous work in this vein includes agent time horizons, a commonly-used metric…