--- title: 'Senior Research Scientist/Engineer, Deception at FAR.AI' canonical: 'https://feeny.ai/job/senior-research-scientist-engineer-deception-far-ai-berkeley-chhrjc0s86e9' type: 'job' last_seen: '2026-11-04' --- # Senior Research Scientist/Engineer, Deception at FAR.AI - **Company:** FAR.AI - **Location:** Berkeley - **Compensation:** $225k–$350k - **Employment:** full-time - **Work type:** onsite - **Posted:** 2026-11-03 - **Last confirmed live:** 2026-11-04 - **Apply:** https://jobs.ashbyhq.com/far.ai/a9146902-8d64-4324-9eeb-ce648944bcf4 ## Job description ## About Us [FAR.AI](http://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. Since our founding in July 2022, we've grown to [50+ staff](https://www.far.ai/about/team), published [40+ academic papers](https://scholar.google.com/citations?user=FVJ24k8AAAAJ), and convened leading [AI safety events](https://far.ai/events/). Our work is recognized globally, with publications at premier venues such as NeurIPS, ICML including a [Best Paper Honorable Mention in 2026](https://icml.cc/virtual/2026/oral/71065), and ICLR, and features in the [Financial Times](https://www.ft.com/content/175e5314-a7f7-4741-a786-273219f433a1), [Nature News](https://www.nature.com/articles/d41586-024-02218-7), [Wired Magazine](https://www.wired.com/story/jailbreaking-ai-models-google-anthropic-openai-spacexai/) and [MIT Technology Review](https://www.technologyreview.com/2020/02/28/905615/reinforcement-learning-adversarial-attack-gaming-ai-deepmind-alphazero-selfdriving-cars/). We conduct pre-deployment testing on behalf of frontier developers such as OpenAI, independent evaluations for governments [including the EU AI Office](https://www.far.ai/news/far-ai-selected-to-lead-eu-ai-act-cbrn-risk-consortium), and publish the [AI Security Leaderboard](https://leaderboard.far.ai/) based on our red-teaming expertise. We help steer and grow the AI safety field through [developing](https://arxiv.org/abs/2405.06624) [research](https://arxiv.org/abs/2506.20702) [roadmaps](https://www.researchgate.net/publication/396910034_Open_Technical_Problems_in_Open-Weight_AI_Model_Risk_Management) with renowned researchers such as Yoshua Bengio; running [FAR.Labs](https://www.far.ai/programs/far-labs), an AI safety-focused co-working space in Berkeley housing 40+ members; and supporting the community through [targeted grants](https://www.far.ai/programs/grantmaking) to technical researchers. ## About the team We explore promising research directions in AI safety and scale up only those showing a high potential for impact. When an approach proves effective, we develop it into a minimum viable demonstration and work with AI developers and governments to support real-world adoption. Our recent and ongoing research includes: - Mitigating AI deception: studying when [lie detectors induce honesty or evasion](https://www.far.ai/news/avoiding-ai-deception), and developing [approaches](https://www.far.ai/research/the-obfuscation-atlas-mapping-where-honesty-emerges-in-rlvr-with-deception-probes) to [state of the art falsehood detectors](https://arxiv.org/abs/2610.12445) and sandbagging. - Red-teaming: conducting pre- and post-release adversarial evaluations of frontier models (e.g. [Claude 4 Opus](https://x.com/ARGleave/status/1926138376509440433), [ChatGPT Agent](https://cdn.openai.com/pdf/839e66fc-602c-48bf-81d3-b21eacc3459d/chatgpt_agent_system_card.pdf), [GPT-5](https://cdn.openai.com/gpt-5-system-card.pdf)); developing [novel attacks](https://www.far.ai/news/defense-in-depth) to support this work. - Evals: developing evaluations for new threat models, e.g. [persuasion](https://arxiv.org/abs/2506.02873) and [tampering risks](https://arxiv.org/abs/2507.11630), and launching a new research agenda on eval awareness - Applied Interpretability: using interpretability to tackle concrete safety problems (better probes, backdoor detection, deception monitoring), aiming for fast feedback loops, often in collaboration with our other pods. - Adversarial Robustness: working to rigorously solve security problems through building a science of security and robustness for AI, from [demonstrating superhuman systems can be vulnerable](https://far.ai/post/2023-07-superhuman-go-ais/), to [scaling laws for robustness](https://www.far.ai/news/does-robustness-improve-with-scale) and [jailbreaking constitutional classifiers](https://arxiv.org/abs/2506.24068). - Mechanistic Interpretability: [finding](https://arxiv.org/abs/2502.12892) [issues](https://arxiv.org/abs/2508.16560) [with](https://arxiv.org/abs/2505.11756) Sparse Autoencoders, probing deception using [AmongUs](https://arxiv.org/abs/2504.04072), understanding [learned planning](https://far.ai/post/2024-07-learned-planners/) in SokoBan, and interpretable data attribution. ## About the Role We are looking for an experienced research scientist/engineer to partner with the Research Lead of the Deception team. You’ll lead research projects from initial proposal to implementation, with scope to contribute across multiple projects and teams. The team works on the following directions: - Detecting deception. Developing white-box probes to identify deception from internal activations, assess failure modes, compare with black-box monitors, and study the effects of self-rationalization. - Monitoring deployed models. Turning probes into real-time monitors with low false-positive rates, minimal overhead, and reliability across tasks and model versions. - Training for honesty. Using probes as a training signal and testing whether models become more honest or simply better at concealing deception. We’ve shown results on an experimental scale, and the next step is scaling them up to a level where frontier labs can adopt them. In addition to providing research judgment, you’ll lead that scaling work and help turn results into methods other teams can use. ## What you'll do - Lead research projects on various timescales on white-box deception detection, probe-based monitoring, and training against probes. Responsibilities include defining research questions, setting milestones, choosing experiments, and assessing the quality of results. - Run post-training and RL experiments on large models, including RLVR and fine-tuning with probe-based penalties; diagnose training instability, reward hacking, and infrastructure failures. - Design model organisms that stress-test methods against realistic failure cases. - Mentor Members of Technical Staff to strengthen research judgment and experimental rigor. - Partner with the infrastructure team to plan and run large post-training jobs on [FAR.AI](http://FAR.AI)’s compute cluster. - Enable frontier labs to adopt the team’s methods through reproducible code, clear evaluations, practical guidance, and direct collaboration with lab safety teams. - Write and present findings for research papers, frontier lab partners, and the broader AI safety community. ## Requirements - 3+ years experience with implementing post-training, model training, or closely related efforts in a frontier AI environment - Hands-on experience post-training models at scale (RL, RLHF/RLVR or large fine-tuning runs), ideally on models of 400B+ parameters. - Strong Python and PyTorch, and experience working in large, shared research codebases. - Strong research judgement to know what experiments to run next and to recognize when a result needs further scrutiny. - Based in or willing to work in U.S. time zones. In person in Berkeley is preferred. If you meet most of this and not all of it, we encourage you to apply anyway. Additional skills we're excited about Real depth in one or more of these makes you a compelling candidate: - Safety or alignment research at a frontier lab or a research lab working on large models (1T+ parameters) - Experience with interpretability or white-box methods (e.g probes, steering or SAEs) - Experience building model organisms of misalignment, or studying reward hacking, scheming or evaluation awareness. - Publications at top ML venues (NeurIPS, ICML, ICLR) or widely cited technical reports. - A PhD in a relevant field. Logistics If based in the USA or Singapore, you will be an employee of [FAR.AI](http://FAR.AI) (501(c)(3) research non-profit / non-profit CLG). Outside the USA or Singapore, you will be employed via an EOR organisation on behalf of [FAR.AI](http://FAR.AI) or as a contractor. - Location: Both remote within U.S. time zones and in-person (Berkeley, CA) are possible. We sponsor visas for in-person employees. - Hours: Full-time (40 hours/week). - Application process: 1 short intake call with a technical recruiter, 1 30 minute screening call with the Head of Engineering, a 60 min ML assessment, a 60 min code review interview, a team fit assessment with the hiring manager, and a paid work trial lasting up to 1 week. If you are not available for a work trial we may be able to find alternative ways of testing your fit. If you have any questions about the role, feel free to contact us 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. Please don't email us to share your resume (it won't have any impact on our decision). Thank you! ## About FAR.AI ## Company Overview - **One-liner**: FAR.AI is a technical AI safety research non-profit dedicated to ensuring advanced AI systems are safe and beneficial for everyone through in-house research, grantmaking, and global coordination events. - **Entity Type**: Private (Non-profit, fiscally sponsored project) - **Headquarters**: Berkeley, California, USA - **Founded**: July 2022 (incorporated October 2022) - **Founders**: Adam Gleave (CEO) and Karl Berzins (President) ## Core Business - **Primary industry**: AI Safety Research, Research Services - **Target customers**: Policymakers, industry leaders, academic researchers, and the broader AI safety ecosystem (primarily B2B/Institutional) - **Mission**: To ensure advanced AI is safe and beneficial for everyone. The organization is motivated by the potential risks posed by rapid advances in AI capabilities. ## Products & Services - **FAR.Research**: In-house technical research team exploring early-stage agendas for AI safety, including adversarial robustness, AI control, and scaling issues. Publishes influential papers and open-source tools. - **FAR.Labs**: A collaborative co-working space in Berkeley for AI safety researchers and organizations, hosting over 40 active members and fostering a thriving community. - **FAR.Futures (formerly FAR.Grants)**: A targeted grantmaking program supporting academics and independent researchers developing innovative solutions to critical AI risks. Has directed millions of dollars in funding. - **Events & Conferences**: Organizes high-impact events including the Alignment Workshop series (global), Berkeley ControlConf 2026 (AI control), and the Technical Innovations for AI Policy (TIAP) Conference, connecting policymakers with leading AI technical experts. ## Market Standing - **Valuation/Market Cap**: Not applicable (non-profit) - **Key Metric**: Total Funding — Not disclosed (non-profit, fiscally sponsored). Key output: 30+ research publications, 1000+ attendees across 10+ events. - **Notable Investors/Partners**: Collaborations with leading think tanks, academic groups (e.g., CHAI, MATS, Apart Research), and major media outlets (Nature, MIT Technology Review, Financial Times, NYT). Fiscally sponsored project of IDAIS. - **Growth Signals**: Headcount grew 53.1% YoY (from ~17 to 34 employees as of mid-2026). Expanded to 8 countries (US, UK, Germany, Spain, Switzerland, Singapore, Mexico, Australia). LinkedIn followers grew 171.4% yearly. Research cited in US Senate hearings (Stuart Russell testimony). ## Competitive Advantages - **Technical Breakthroughs**: Published influential work on adversarial attacks on superhuman Go AIs (featured in Nature), multi-agent adversarial policies, and red teaming leading language models for frontier labs. - **Ecosystem Centrality**: Runs the only dedicated AI safety coworking space (FAR.Labs) and a premier grantmaking program, positioning FAR.AI as a hub connecting academia, industry, and policy. - **Policy Influence**: TIAP Conference directly connects policymakers with technical experts; research cited in congressional testimony. - **Top Talent Magnet**: Attracts researchers from top AI labs (Anthropic, Google DeepMind) and academic safety hubs (Cambridge AI Safety Hub, CHAI). ## Strategic Focus - **Field Building**: Scaling the global AI safety ecosystem through grants, events, and coworking space. - **Technical Innovation**: Continuing to explore early-stage, high-impact research agendas (e.g., AI control, robustness, adversarial training) until they can be adopted by the broader community. - **Policy Engagement**: Deepening connections between technical experts and policymakers to ensure safety techniques are adopted. - **Talent Development**: Expanding the team with world-class researchers, engineers, and operations staff to tackle critical AI safety challenges. ## Why Work Here - **Mission-Driven Culture**: Employees report high satisfaction (5.0/5.0 on Culture, 5.0/5.0 on Work-Life on LinkedIn). The mission to make advanced AI safe is described as "one of the most critical challenges of our time." - **Work Environment**: Hybrid/remote-first with a physical HQ in Berkeley. Offers both remote and onsite roles. The Berkeley office (FAR.Labs) fosters a collaborative, startup-like atmosphere. - **Team Composition**: Small, high-agency team of ~34 people with a flat structure. Heavy emphasis on research (18% of staff) and project management (12%). Senior leadership includes former Anthropic and DeepMind researchers. - **Compensation & Perks**: Rated 4.0/5.0 on Compensation. As a non-profit, likely offers competitive (but not market-maxing) salaries with strong mission alignment. Perks include the coworking space, events, and a highly collaborative environment. - **Career Growth**: High growth trajectory (53% YoY headcount increase) means rapid advancement opportunities. Employees often move to top AI labs (Anthropic, DeepMind, Stability AI) after their tenure. - **Notable**: Currently hiring for Jailbreaking Lead (Red Team), Executive Operations Assistant, and Technical Project Manager (Red Team). ## Sources 1. [far.ai](https://www.far.ai/) 2. [far.ai/about](https://www.far.ai/about) 3. [far.ai/careers](https://www.far.ai/careers) 4. [linkedin.com/company/far-ai](https://www.linkedin.com/company/far-ai) 5. [builtin.com/company/farai](https://builtin.com/company/farai) ## Other roles at FAR.AI - [Senior Media Manager](https://feeny.ai/job/senior-media-manager-far-ai-united-states-p4v5a99atday) — United States - [Community Lead](https://feeny.ai/job/community-lead-far-ai-berkeley-b5sx3wh9kczn) — Berkeley - [Tech Lead Manager, GPU Cluster Infrastructure](https://feeny.ai/job/tech-lead-manager-gpu-cluster-infrastructure-far-ai-global-qcbffm40ehej) — Global - [Senior Software Engineer, GPU Cluster Infrastructure](https://feeny.ai/job/senior-software-engineer-gpu-cluster-infrastructure-far-ai-global-ps8k924d9yph) — Global - [Research Lead](https://feeny.ai/job/research-lead-far-ai-berkeley-x59jybp60rbj) — Berkeley - [Research Scientist, Applied White-Box Methods](https://feeny.ai/job/research-scientist-applied-white-box-methods-far-ai-berkeley-t5q02qecrqa9) — Berkeley - [Head of Engineering (Red Team)](https://feeny.ai/job/head-of-engineering-red-team-far-ai-global-qg9gyxwf3674) — Global - [Staff Accountant](https://feeny.ai/job/staff-accountant-far-ai-berkeley-878dyjxya3j3) — Berkeley - [IT & Security Manager](https://feeny.ai/job/it-security-manager-far-ai-united-states-gn70wqjc2gbh) — United States - [Research Lead - Pre-training Safety](https://feeny.ai/job/research-lead-pre-training-safety-far-ai-berkeley-z2xbmrwva3e6) — Berkeley