Security Labs Engineer

Details

Anthropic’s mission is to create reliable, interpretable, and steerable AI systems. The Security team treats security as a core part of building increasingly capable systems and runs structured security R&D projects (6-month experiments) to resolve high-uncertainty questions about long-term security posture.

About the Role

As a Security Labs Engineer you will own one or more security R&D projects end-to-end: scoping experiments, building prototypes and infrastructure, coordinating across teams, running pilots, documenting results, and driving successful experiments toward production scale. Work is 0-to-1 and 1-to-10, with a focus on rapid learning and measurable exit criteria.

Current Project Areas

• Designing and operating a mock high-assurance research environment (extreme isolation and physical security controls) and measuring productivity impact
• Exploring cryptographic verification of model integrity (e.g., zero-knowledge proofs) to provide mathematical guarantees about production workloads
• Assessing feasibility of confidential computing across the model lifecycle
• Piloting AI-assisted security tooling: vulnerability discovery, automated patching, anomaly detection, adaptive behavioral monitoring
• Prototyping API-only access regimes to avoid exposing raw model weights to internal workflows

Responsibilities

• Own end-to-end execution of Security Labs projects: refine hypotheses, design experiments, build prototypes, run pilots, and write up results
• Build novel security infrastructure under time pressure: isolated clusters, hardened access controls, cryptographic verification layers, with a bias toward learning fast
• Drive successful experiments toward production scale (ensure solutions work beyond a single cluster)
• Work embedded with research teams (Pretraining, RL, Inference) to stress-test core workflows under extreme security controls and document breakage points
• Evaluate and integrate emerging security technologies through coordination with vendors and research groups
• Produce clear, decision-ready writeups that inform long-term security architecture and Responsible Scaling Policy (RSP) commitments
• Maintain a pain-point registry and feasibility assessment for each project and help scope/prioritize future Labs projects

Requirements

• 7+ years of software or security engineering experience with a solid foundation in production systems
• Experience building pilots, prototypes, or applied research efforts where shipping a working answer to a hard question was the explicit goal
• Strong programming skills in Python and at least one systems language (Go, Rust, or C/C++)
• Hands-on experience with cloud infrastructure (AWS, GCP, or Azure), Kubernetes, and networking fundamentals sufficient to stand up and tear down isolated environments quickly
• Track record of cross-functional execution with ML researchers, infrastructure engineers, and vendors
• Clear written communication: ability to turn weeks of experimentation into concise, actionable memos
• Comfort with ambiguity and iteration; experience running experiments that failed and extracting lessons
• Genuine curiosity about defending against high-capability (nation-state-level) adversaries
• Passion for AI safety and understanding of the role security plays in frontier AI development
• Bachelor's degree in Computer Science, a related field, or equivalent industry experience (required)

Strong Candidates May Also Have

• Experience in offensive security, red teaming, or security research
• Familiarity with airgapped or high-side environments (classified networks, ICS/SCADA, financial trading infrastructure, or similar)
• Knowledge of applied cryptography: zero-knowledge proofs, attestation protocols, secure enclaves, TPMs, or confidential computing primitives
• Experience with ML infrastructure (training pipelines, inference serving, model packaging)
• Background in fast-iteration security systems, startups, innovation teams, or applied research groups

Location and Office Policy

This role is based in Anthropic’s San Francisco office (500 Howard St). Several Labs projects involve physical secure facilities on-site, so you should expect to be in-office more frequently than Anthropic's standard 25% hybrid baseline.

Compensation

Annual Salary: $320,000 - $405,000 USD

Logistics

• Education requirements: at least a Bachelor's degree in a related field or equivalent experience
• Location-based hybrid policy: staff expected to be in one of Anthropic's offices at least 25% of the time; some roles may require more in-office presence
• Visa sponsorship: Anthropic does sponsor visas and retains an immigration lawyer to assist, though sponsorship may not be possible for every role/candidate

How We're Different

Anthropic pursues large-scale research efforts as a cohesive team, values communication, and emphasizes empirical science approaches to AI research. The team prioritizes impact on long-term goals (steerable, trustworthy AI) and frequent cross-disciplinary research discussions.

Benefits

Anthropic offers competitive compensation and benefits, optional equity donation matching, generous vacation and parental leave, flexible working hours, and a collaborative office environment.