See how candidates think and approach the work this role demands, before the phone screen. We'll build a video challenge from this posting, and you can edit or share it before it goes live.
A frontier model lab is staffing security for its own employees' use of AI agents and third-party tools, signaling that the risk surface now includes staff delegating work to autonomous agents, not just protecting models and infrastructure.
We’re the team building the foundational layer for visual intelligence: multimodal models trained jointly on images, video, audio, and action to see, generate, and understand the world.
Our founding team pioneered latent diffusion and Stable Diffusion, the breakthroughs that made generative AI accessible to millions. With FLUX, we're pushing into generative media at scale, machines that act in the physical world, and emerging applications in computer use.
Our FLUX models top public benchmarks and run in production in some of the world's biggest platforms. We're a small, fast growing research lab competing with the largest companies in the world. From our labs in Freiburg to San Francisco, we're building what comes next.
We develop AI models and run the infrastructure on which those models are trained, deployed, and served to customers. We're looking for an experienced, hands-on security engineer to protect how our staff work: the accounts they sign in with, the devices they use, the third-party tools they rely on, and the agents they delegate work to. You'll work closely with IT, which operates our employee services and devices.
This is an engineering role: you'll spend your time building tools and automations, designing practical controls, investigating escalations, and shipping changes.
These are the areas this role draws from; you're not expected to cover
them all at once:
Identity and access. Strengthen how staff authenticate and what they can reach: single sign-on coverage, phishing-resistant authentication, and access provisioning and revocation that stay reliable and complete as people and tools change. Design scoped, short-lived credentials for the tools and agents that act on staff's behalf, and improve our ability to detect credential misuse.
Endpoint security. Work with IT on device trust, hardening, and visibility across our fleet. Find practical ways for engineers and researchers to install what they need without introducing unnecessary risk.
AI agents and staff tooling. Help staff use agents and AI-powered tools productively and safely. Review the threat models, permissions, and data flows of tools staff build or adopt, including coding agents, desktop assistants, browser extensions, and MCP servers. This includes how staff supervise and stay accountable for delegated actions, and the audit trails that make agent activity reviewable. Build sandboxes, credential patterns, and guardrails that make the secure path the convenient one.
Third-party tool security. Define a scalable security review for the tools staff use day to day, and work with IT, which owns procurement and provisioning, to build security checks into how tools are onboarded and offboarded. Maintain a current picture of which tools can reach company code, credentials, and customer data as tools are added and retired.
Visibility and response. Make sure identity, endpoint, and SaaS telemetry is collected and usable. Investigate escalations and incidents in these areas, verify that we can contain a compromised account or device quickly, and turn findings into lasting fixes.
Strong experience in identity and access security and endpoint security. Hands-on experience administering an identity provider and a device-management platform.
Practical understanding of OAuth, OIDC, SAML, secrets handling, and authorization design. You can trace how an application obtains and uses credentials and find where access controls can be bypassed.
Hands-on engineering ability. You can read unfamiliar code, inspect how an app talks to APIs, and write the automation and tooling that replace manual processes.
Experience using LLMs and agents in your own work, along with a clear-eyed view of where they help, where they don't, and how they can be attacked or misused.
Good security judgment. You can distinguish meaningful risks from theoretical ones and apply controls in proportion to the risk.
Experience supporting researchers, engineers, or other highly technical staff, and a track record of finding secure ways to meet their needs.
Comfort working with real autonomy and turning loosely defined problems into concrete improvements.
Experience with macOS security, developer tooling, SaaS security, software distribution, agent integrations, or threat hunting would be particularly relevant, but none is a prerequisite. We're more interested in people who have solved difficult security problems for real users than in candidates who can check every box on a technology list.
We’re a distributed team with real offices that people actually use. This role is based in San Francisco, and you’ll join us in the office at least 2 days a week, or one full week every other week.
Everything we do is grounded in four values:
Obsessed. We are a frontier research lab. The science has to be right, the understanding deep, the product beautiful.
Low Ego. The work speaks. The best idea wins, no matter who said it. Credit is shared. Nobody is above any task.
Bold. We take the ambitious bet. We ship, we do not wait for conditions to be perfect.
Kind. People over politics. We treat each other with genuine warmth. Agency without empathy creates chaos.
If this sounds like work you’d enjoy, we’d love to hear from you.
Base Annual Salary:
US $180,000 - $225,000 USD + Equity
Note: We may use AI-assisted tools to help organize applications. All hiring decisions are always made by our team.
blackforestlabs:92920499-3d5c-422e-9849-1c5f5f7ef5a8See how we measure for definitions, or our corrections log for known issues. Found something wrong? Flag a correction.
black-forest-labsThis posting hasn't been probed by our closure verifier yet. Stream C runs on a rolling schedule against postings approaching the close-decision threshold.