Research Scientist, Applied White-Box Methods

FAR.AI is a non-profit AI research institute working to ensure advanced AI is safe and beneficial for everyone. Our mission is to facilitate breakthrough AI safety research, advance global understanding of AI risks and solutions, and foster a coordinated global response.

We’re structured to support that work from early research through real-world adoption:

Independent by design. We can pursue what's most impactful based on our theory of change and share what we find publicly.

A portfolio approach. Rather than focus on one single direction, we run diverse bets across the safety stack. We take promising ideas from initial experiments to deployment, informed by red-team partnerships with frontier labs and governments.

Serious infrastructure for ambitious research. A dedicated engineering team runs our compute cluster and experiment-scaling stack, so researchers spend their time on research instead of on infra.

Setting the standard. Our events convene key decision makers; our red-team works with frontier developers and governments; and our communications inform the public. Together, this drives adoption and sets the new standard in safety.

Since our founding in July 2022, we've grown to 50+ staff, published 40+ academic papers, and convened leading AI safety events. Our work is recognized globally, with publications at premier venues such as NeurIPS, ICML including a Best Paper Honorable Mention in 2026, and ICLR, and features in the Financial Times, Nature News, Wired Magazine and MIT Technology Review. We conduct pre-deployment testing on behalf of frontier developers such as OpenAI and independent evaluations for governments including the EU AI Office and publish the AI Security Leaderboard based on our red-teaming expertise. We help steer and grow the AI safety field through developing research roadmaps with renowned researchers such as Yoshua Bengio; running FAR.Labs, an AI safety-focused co-working space in Berkeley housing 40+ members; and supporting the community through targeted grants to technical researchers.

The Applied White-Box Methods team develops, evaluates, and demonstrates methods that leverage model internals to improve the safety of AI systems. We work on diverse AI safety applications of white-box methods, from white-box control to evaluation awareness to shaping training dynamics to improve alignment.

Black-box methods, such as chain-of-thought monitoring, work well for now, but we are quickly entering a world where black-box interventions and monitoring are insufficient. Interpretability research is still often early-stage, curiosity-driven work without realistic evaluations on applications that matter. The team bridges the gap between exploration and deployment by stress-testing white-box methods on real-world (e.g. long-context agentic coding) tasks and using this feedback loop to enable the deployment of better white-box methods at frontier scale.

We use the term "white-box" deliberately: our scope includes any method that uses model internals to understand, predict, or intervene on model behavior, not only what is conventionally called interpretability. Methods of current interest include natural language autoencoders and other activation explainers, activation oracles, steering, patching, and other activation-level interventions, influence functions and data attribution, and singular learning theory. We evaluate our methods against real baselines – strong black-box methods and activation probes – to be able to make an honest case that the methods are worth implementing, or conclude that simpler methods work better for now.

AI Research Automation. AI will soon automate most of the hill-climbing in the research process. We anticipate this by focusing our effort and judgement on defining realistic evaluations with Goodhart-resistant metrics. With the evaluation framework properly built, we can pour vast amounts of AI labor into method development and iteration without overfitting. We believe this is the way to scale white-box research into the age of RSI.

Realistic, large-scale models. Most model organisms in interpretability research are narrow SFT on an instruct model. Such unrealistic models lead to unconfident conclusions about which methods do and do not work. We leverage FAR's shared compute and infrastructure to work on organisms that come out of pipelines a frontier lab could plausibly have run: for example, reward seekers trained by RL in broken environments, with other contaminated data mixed in to induce other misalignments.

Realistic, large-scale evaluations. We will primarily study long-context agentic coding as this is where most of the risk currently lies. In addition to the classic AI control sabotage settings, we will also study reward hacking, sandbagging, and research tampering, which are some of the key failure modes that matter during RSI.

Practical monitors and interventions. Deployment of new methods has real costs for AI developers. Part of our focus on real-world applications is ensuring that methods are simple and efficient enough to deploy at frontier scale.

As a Research Scientist on the team, you will take ownership of and accelerate the team's research agenda, publish findings broadly, and engage with the AI alignment community. You are encouraged to propose new directions within the team's agenda. You are welcome and encouraged to attend relevant conferences and other community events, and leverage FAR.AI’s existing comprehensive infrastructure for events convening and government relations as you see fit. Beyond FAR.AI, you can work with national AI safety institutes, frontier model developers, and top academics.

Two backgrounds are particularly well suited to this role: researchers with some experience in more fundamental interpretability who want to evaluate and refine those methods in realistic settings, including agentic coding, long contexts, and realistic threat models; and researchers from evaluations, AI control, red-teaming, or reinforcement learning who have begun working with model internals and want to deepen that work. In both cases, prior hands-on experience with white-box methods and a demonstrated interest in their practical application are preferred. (Note that this role is not well suited to researchers whose primary interest is interpretability as a foundational science.)

If you are new to AI Safety research, that is also ok! But please be prepared to explain how your previous research experience (e.g. in applied ML) could be leveraged for our work, and how you are engaging with the field of technical AI safety research today.

We are hiring across a range of levels of experience. You may already have some of the following, though not necessarily all.

• Hands-on experience applying at least one white-box method to a real model (activation explainers, SAEs, steering, attribution, influence functions, probes, or similar), and an informed view of its limitations.

• A track record in AI safety: a paper, a fellowship project, or substantive public writing. Alumni of programs such as MATS, Astra, Anthropic Fellows, SPAR, Pivotal, LASR or similar programs are especially encouraged to apply.

• Experience with evaluations, AI control, red-teaming, reinforcement learning, or post-training of LLMs. Experience running large training runs is a plus.

• The ability to communicate novel methods and results clearly to technical and non-technical audiences.

• A PhD or several years of research experience in computer science, machine learning, physics, statistics, or a related field.

• Previous experience in applied ML for other fields: e.g. biology, chemistry, materials science, robotics, etc.

• Health Insurance - 94% of Insurance premium paid by Organization commencing within 1 month after your start date

• Retirement - 401(k) plan with up to 2% match

• PTO - 25 days Paid Time Off per year, accrued weekly and up to 10 days of paid sick leave per year

• Paid Leave - Paid Bereavement, Family, Medical and Pregnancy Disability Leave

• WFH Stipend & Equipment - Work computer and stipend provided for eligible employees

• Catered Meals (Berkeley Office Only) - Catered lunches and dinners on workdays at our office

*(Available only to full-time employees located in the US)

• If based in the USA, you will be an employee of FAR.AI, a 501(c)(3) research non-profit. Outside the USA, you will be an employee of an EoR organization on behalf of FAR.AI.

• Location: Berkeley, CA. This role is in-person; we strongly prefer candidates who are in the Bay Area or willing to relocate, and we sponsor visas for in-person employees. We will consider remote arrangements for exceptional candidates.

• Hours: full-time (40 hours/week).

• Compensation: $150,000–$250,000/year depending on experience and location, plus work-related travel and equipment expenses. Catered lunch and dinner at our Berkeley office.

• Application process: interviews with members of our technical staff, followed by a paid work trial of up to one week. If you are not available for a work trial we may be able to find alternative ways of assessing fit.

If you have any questions about the role, feel free to contact us at [email protected]. Otherwise, if you don't have questions, the best way to ensure a proper review of your skills and qualifications is by applying directly via the application form. Please don't email us to share your resume (it won't have any impact on our decision). Thank you!