IT & Security Manager
FAR AI
IT & Security Manager
FAR AI
Non-profit AI safety research institute
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, 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, 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.
We are hiring an IT & Security Manager to keep FAR.AI running reliably and securely as we grow. This is a hands-on role for someone who thinks about internal systems the way a security-minded engineer does: curious about what an attacker would actually go after, and more interested in systems that hold up under pressure than in systems that pass an audit.
You will report to the COO and work closely with our Head of Research Acceleration, who owns research compute and cluster infrastructure. You own everything on the organizational side: accounts, devices, identity and access, the FAR.Labs’ office network and the security posture underneath it. In the short- to mid-term future, you will hire an in-person IT Coordinator to handle day-to-day support, and you will manage them. You are the first person on this team, and you will shape what it becomes.
IT at FAR.AI has been distributed across finance, business operations, research, and the executive team since our founding. That worked when we were 15 people, but has become a barrier at 50. We’re intentionally not looking for an MSP, and instead looking for someone we’re excited to build alongside.
Although our security posture is reasonably strong, we are looking to improve this to withstand targeted AI driven attacks as we increasingly work in security critical areas. We have a documented list of known improvements waiting for an owner, a risk surface that grows with every frontier lab partnership, and researchers who move fast and need to keep moving. You will be working on real problems in your first week, with a mandate to go after root causes.
IT operations: Own our core internal systems end-to-end: accounts, devices, identity and access management, onboarding and offboarding, the FAR.Labs office network, and the daily support that keeps a growing team working.
Security remediation: Work through our known security gaps, prioritized by risk. Access controls, endpoint hardening, data protection, data governance, and vendor and application risk are the near-term substance.
Self-serve infrastructure: Reduce the amount of work that has to route through you. Partner with Business Operations and other teams to design systems and processes that reduce manual work.
Cross-functional partnership: Work with team leads across Research, Operations, Futures (events), and Communications to understand how they actually work and help them build securely inside their own domains. You work cross-functionally to drive AI adoption.
Building the team: Hire the in-person IT Coordinator, then manage and develop them. You get to define the role in accordance with the organization’s needs. Over time, this team may grow.
Internal communication and change management: Roll out changes in a way people can live with. You communicate clearly about what is changing and why, and establish feedback loops to continue iterating post-deployment.
You have owned IT and security at a technical organization of roughly 30 to 150 people, ideally as the first or sole person to do so. You built the function yourself, without a team already in place around you.
You are comfortable in a terminal and can code. You do not need to be an excellent engineer, but you are not afraid of it. You reach for automation before a manual process.
You think about the people using what you build. You go after root causes instead of managing symptoms; you can point to a system you redesigned so that a whole class of requests stopped arriving, and you treat that as ongoing work rather than a one-time project.
You have done hands-on security work driven by threat models. You can say who you were defending against and what you chose not to spend effort on, not just which frameworks you mapped to.
You have implemented and owned core identity and device infrastructure, including MDM and SSO.
You are fast and service-oriented. You read the support queue as information about the system, and the people you have supported would say you were reliable.
You communicate clearly across technical and non-technical audiences. You can point to a time when you rolled out something that made someone's day harder, and you brought them along anyway.
You use LLMs regularly in your own work and have an opinion on where they help and where they do not.
You care about AI safety or security, or you are demonstrably curious about it. You understand why this work matters and you help assess the value of the risk we’re taking on.
You have hired, managed, or mentored a junior teammate.
You engage with security as a craft outside of work: your own compute build or a home lab, capture-the-flag competitions, side projects, or conferences like DEF CON or CCC.
You have experience obtaining SOC 2 Type 2 and complying with GDPR. We are open to contracting out much of the work in obtaining certifications – our primary goal is to improve our actual security posture, not achieve certifications, but they may be a necessary step in certain engagements.
You have experience in fast-paced environments. You've worked in startups, nimble nonprofits, or high-growth teams before and can adapt to change quickly.
You have worked in a research organization, a deep-tech startup, or a similarly technical environment.
Compensation: $185,000 to $250,000 per year. Compensation is determined by our salary bands and your level of experience, with adjustments for location. We also pay for work-related travel and equipment expenses, and offer catered lunch and dinner at our office in Berkeley.
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
*(Available only to full-time employees located in the US)
Test task (we pay for 90 minutes of your time, and most people finish in less),
Initial screen with a Talent Partner
Interview with the hiring manager
Interview with our Head of Research Acceleration
A full day with the team, either in Berkeley or remote, that includes a working project and conversations across our operations and leadership team. *
*Some candidates will have a short final conversation after that.
References
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!
FAR AI is a non-profit research institute focused on making advanced AI safe and beneficial for society. It conducts in-house work on model evaluation, interpretability, and robustness, and it funds collaborations to pursue safety research ideas. A core activity is red-teaming frontier AI systems to uncover vulnerabilities and inform safety standards for developers and governments. The organization runs FAR.Labs in Berkeley and hosts global dialogues to foster collaboration, supported by philanthropic funding with for-profit revenue kept to 10% to preserve independence.
Explore more opportunities at FAR AI