FAR.AI

Role Description Jailbreaking is the core technical engine of the red team. As Jailbreaking Lead, you own that engine. You are the person who personally breaks the hardest targets, sets the bar the rest of the team pushes toward, and makes sure we keep discovering the highest severity, universal vulnerabilities – the most important vulnerabilities to fix – in the most heavily defended frontier models on the planet, faster than anyone else. We expect you to spend at least 50-70% of your time hands-on across 2026: - Breaking models, chaining novel attack classes through defense-in-depth stacks. - Helping to invent new techniques when existing ones fail. - Setting the standard for what constitutes a significant vulnerability and a credible mitigation. The remaining time will go to managing/mentoring ICs, helping to shape the jailbreaking research agenda with Kellin, and making sure our findings land with frontier labs, governments, and the broader field. This is a senior IC role by default, intended to attract a world-class jailbreaker whose personal mission is to find critical jailbreaks in the most heavily defended domains of the leading frontier AI models, and who has a track record of repeatedly doing so. In practice, this role spans: - Lead jailbreaking on the highest-stakes engagements: - Personally develop universal and near-universal jailbreaks against frontier closed- and open-weight models. - Systematically dismantle defense-in-depth stacks. - Escalate initial vulnerabilities to expose their most severe form. - Own the technical bar for vulnerability severity and generality on every major engagement. - Push the frontier of jailbreaking techniques: - Invent new attack classes when existing techniques fail. - Monitor and rapidly incorporate state-of-the-art methods from the literature. - Shape the jailbreaking research agenda in partnership with Kellin. - Stress-test novel affordances as frontier systems evolve. - Raise the technical bar across the team: - Set the standard for rigour, creativity, and precision in jailbreaking across the red team. - Mentor ICs on attack craft. - Review major red-teaming deliverables for technical quality. - If on the management track: hire, manage, and grow a jailbreaking team. - Translate jailbreaks into real-world impact: - Work directly with frontier labs and government agencies. - Contribute to public reports, benchmarks, and the FAR.AI safety leaderboard. - Make precise, calibrated technical judgments about vulnerabilities. Qualifications - Obsess over frontier model jailbreaks. - Track record of finding non-obvious, high-severity vulnerabilities in frontier AI systems. - Combine deep technical craft with the judgment to know which vulnerabilities matter. - Excited by high-stakes, real-world technical work. - Value independence and the ability to publish and speak honestly about risks. - Care deeply about AI safety. - Has a “get shit done” attitude. - Thrive in fast-moving, ambiguous environments. Requirements - Personally developed universal or near-universal jailbreaks against at least one leading frontier model. - Demonstrated ability to discover non-obvious, high-severity vulnerabilities. - Deep, hands-on jailbreaking experience with demonstrated success against modern frontier models. - Experience with black-box optimisation methods, multimodal attacks, and/or agentic red-teaming. - Deep understanding of large language model architectures and failure modes. - Strong existing track record in AI, adversarial ML, security, or another highly technical subject. - Invented novel attack classes. - Demonstrated drive for mission/impact. - Demonstrated relentlessness in achieving ambitious goals. Benefits - Location: Remote globally. We can sponsor US or Singapore visas. - Hours: Full-time. Expect up to one trip per month for convenings, government meetings, or team gatherings. - Compensation: USD 170,000–250,000, depending on experience. Exceptional candidates may be offered more.

We're hiring a Business Operations Manager to join FAR.AI’s BizOps team as the right hand to our Head of Business Operations. You’ll report to the Head of Business Operations and work closely with peers across Finance, People Operations, and Programs. BizOps is the connective tissue of FAR.AI – we design and maintain the systems that enable functional teams to focus on delivery. We own the problems that cut across the organization, fill gaps where there's no clear owner, and act as accelerants when teams need operational support to move faster. This role comes with real ownership and autonomy at a key moment in our growth. As the BizOps Manager at FAR.AI, you'll help shape how we operate and scale. You will be responsible for defining problems, developing solutions, and driving these solutions to completion, maintaining ownership over the long term. While you'll start as an individual contributor, there's room to grow; whether that's into people management, expanded scope, or deeper specialization as the BizOps function evolves. Design and maintain cross-team coordination systems: project trackers, operational dashboards, and workflow tools that support visibility and decision making across FAR.AI. Manage FAR.AI’s operational requests pipeline – triaging cross-team support needs, assessing resource implications, and ensuring requests are scoped, prioritized, and resolved efficiently. Coordinate cross-functional initiatives that require alignment across multiple teams, managing dependencies, timelines, and stakeholder communication. Own operational projects end-to-end, from scoping through implementation (e.g. building a new grant tracking system, or re-designing onboarding workflows with team leads), or driving specific pieces of larger initiatives. Evaluate, optimize, and maintain FAR.AI’s operational tool stack and knowledge management systems, identifying opportunities for process automation and systems integration as FAR.AI scales. Support organizational processes like onboarding, quarterly retrospectives, and recurring operational rhythms. Spot operational risks, process gaps, and emerging compliance requirements, and propose solutions to leadership. Drive continuous improvement across FAR.AI’s operations – identifying inefficiencies, eliminating unnecessary complexity, and redesigning systems so the organization can scale with clarity, speed, and control. Coordinate FAR.AI's grantmaking operations across the full lifecycle: from grant review and due diligence through disbursement, monitoring and evaluation, designing the systems and processes that support each stage, and ensuring compliance with internal policies and nonprofit regulations. Own end-to-end grant compliance across funds FAR.AI receives, managing funder reporting workflows and deadline coordination across internal teams. Coordinate organization-wide compliance processes, including documentation, record-keeping, regulatory filings, and deadline management, in partnership with Finance, People Operations, and external legal counsel. Support compliance-related research across international and domestic operations, flagging regulatory ambiguities to leadership and legal counsel.

Head of Engineering (Research) FAR.AI is seeking a Head of Engineering (Research) to scale one of the world's leading AI safety research teams through a critical growth phase. You will instill excellent people and project management systems to enable our team to rapidly test and iterate on new ideas while maintaining high quality. You will scale the team from three to five research pods, expanding from 15 to 30 technical staff. You will be the key glue that streamlines internal partnerships with other FAR.AI divisions, and manage key relationships with external stakeholders we work with including leading AI companies and government AI safety institutes. We expect this role’s primary focus to be building, managing, and scaling (90%) the technical team with minimal individual contributor (IC) or engineering focused work (10%). This role would be perfect for an experienced technical leader looking to work in an impact-driven environment. About Us FAR.AI is a non-profit AI research institute dedicated to ensuring 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. Since our founding in July 2022, we've grown quickly to 30+ staff , producing over 40 influential academic papers , and establishing leading AI Safety events . Our work is recognized globally, with publications at premier venues such as NeurIPS, ICML, and ICLR, and features in the Financial Times , Nature News and MIT Technology Review . We drive practical change through red-teaming with frontier model developers and government institutes. Most recently, we discovered major issues with Anthropic’s latest model the same day it was released , and worked with OpenAI to safeguard their latest model. Additionally, 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. About FAR.Research Our research team likes to move fast. We explore promising research directions in AI safety and scale up only those showing a high potential for impact. Unlike other AI safety labs that take a bet on a single research direction, FAR.AI aims to pursue a diverse portfolio of projects. Our model is to conduct initial investigations into a range of high-potential areas. We incubate the most promising directions through a combination of in-house research, field-building events, and targeted grants. Once the core research problems are solved, we work to scale them to a minimum viable prototype, demonstrating their validity to AI companies and governments to drive adoption. Our current focus areas include: Mitigating AI deception : studying when lie detectors induce honesty or evasion , and developing model organisms for deception and sandbagging Evaluating and red-teaming : Conducting pre- and post-release adversarial evaluations of frontier models (e.g. Claude 4 Opus , ChatGPT Agent , GPT-5 ); developing novel attacks to support this work; and exploring new threat models (e.g. persuasion , tampering risks ). Robustness : working to rigorously solve these security problems through building a science of security and robustness for AI, from demonstrating superhuman systems can be vulnerable through to scaling laws for robustness . Explainability : developing foundational techniques such as codebook features and AC/DC, and applying them to understand core safety problems like learned planning . FAR.AI is one of the largest independent AI safety research institutes, and is rapidly growing with the goal of diversifying and deepening our research portfolio. For that reason, we’re seeking a Head of Engineering (Research) to manage the team, accelerate delivery, and grow the team to continue to scale our impact. About the Role This role is critical to FAR.AI's mission. Our research directly influences how the most capable AI systems are developed and deployed. You will determine whether we can scale fast enough to keep safety ahead of rapidly advancing AI capabilities. Beyond scaling, you will work directly with leading researchers at FAR.AI to increase the velocity of their research, and with key decisionmakers in industry and government to translate our research to tangible impact. As the Head of Engineering (Research), you will manage FAR.AI’s research team, being responsible for our team’s delivery and execution. You will: Manage the leads of our research pods and engineering pod, holding them accountable for their respective pod’s progress. Instill excellent people and project management systems that enable researchers to iterate rapidly while maintaining rigor, including through mentorship and coaching of pod leads. Scale FAR.Research, growing our technical team from 15 to 30 FTE in the next 12 to 18 months and overseeing our hiring process. Proactively identify bottlenecks before they slow research down, and rapidly design and implement solutions – whether that's streamlining hiring, improving team processes, or connecting researchers with the support they need. Streamline interactions between FAR.Research and the rest of FAR.AI, enabling our communications and operations team to better support FAR.Research, and our team to develop strategic communication, events, and grantmaking initiatives. Ensure our work has real-world impact through deepening our relationships with leading AI companies (e.g. red-teaming) and government departments (e.g. threat assessment, research contracts). Mentor individuals doing deep technical work; however, we expect this role to have minimal IC or engineering focused work. This role would be a great fit if you: Love helping talented individuals reach their full potential. Enjoy working with mission-driven colleagues to achieve real-world impact. Like being in an intellectually curious environment. Enjoy building and scaling high-impact and high-performing teams Want to work with a range of governments, leading AI companies, and academics. We’re a lean organization, and seek to leverage our impact through strategic partnerships. This role would be a poor fit if you: Want to lead a research direction or make direct technical contributions. This is primarily a management role – but do check out our other more technical roles . Want to do significant amounts of engineering in your day to day work. Wish to operate within clearly defined bounds. We’re planning to double in size in the next 12-18 months, and will change a lot as an organization. There will be no shortage of opportunities but navigating uncertainty is a core part of the role. Want an equity upside, or value compensation above all else. We’re a non-profit: we aim to pay competitively, but if this is the most important factor for you, there are other organisations better suited to your priorities. About You To be a strong candidate for the Head of Engineering (Research) role, you likely: Have at least 3 years experience managing technical teams, including some experience managing managers. We expect most candidates will have more experience than this (typically 5+ years, often 10+) but we value a track record of impactful outcomes and potential above simply number of years. Have thrived in rapidly evolving environments and successfully executed on pivots, such as in a start-up, rapidly growing team at a larger company, or in a research environment. Bring technical intuition from past hands-on work as an individual contributor. It’s OK if you’re rusty or the experience was outside AI: you don't need to be actively coding, but you should understand how technical projects unfold and be able to read one of our research papers or design documents, understand the technical approach, and grasp what might be slowing down progress. Can effectively communicate novel methods and solutions to both technical and non-technical audiences. Have an interest in making AI trustworthy and secure. It is preferable if you: Designed and refined hiring processes for technical positions. Have some prior experience operating in a research environment. Have some familiarity working with large language models. Logistics 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 organisation on behalf of FAR.AI. Location : Both remote (global, at least 3-5 hours of overlap with Pacific time) and in-person (Berkeley, CA) are possible. We sponsor US visas for US-based employees, and can also hire remotely in most countries. Hours : Full-time (40 hours/week). Compensation : $200,000-$350,000/year depending on experience and location, with the potential for additional compensation for exceptional candidates. We will also pay for work-related travel and equipment expenses. We offer catered lunch and dinner at our offices in Berkeley. Application process : Apply by submitting a CV and a one paragraph description of why you’re interested in working with us. From there, you’ll be invited to a short phone screen to assess mutual fit, a paid test task to provide feedback on some of our past work, three one-hour interviews covering management, technical mentorship and your technical background; and a paid full-day interview. We know these roles are rare and the skill combination is unusual. If you're uncertain whether your background fits but are excited by the mission and challenges, we encourage you to apply – we're looking for excellence and potential, not a perfect resume match. If you have any questions about the role, please do get in touch at talent@far.ai . 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.