2 days ago
London, United KingdomMid Level
Responsibilities
- Design, train, and evaluate real-time classifiers for detecting unsafe actions by internal agents.
- Deploy, tune, and operate classifiers in the critical path of internal agents, reinforcement learning training, and evaluations.
- Own classifier accuracy, reliability, latency, cost, and failure behavior.
- Build datasets, labeling pipelines, evaluations, and harm taxonomies to measure and improve classifier performance.
- Partner with red-teaming colleagues to stress-test and harden classifiers against adversarial behavior.
- Work with agent, environment, and evaluation teams to resolve false positives while maintaining protection.
- Contribute to team direction, technical priorities, partnerships, and investment decisions.
Requirements
- Organizational, prioritization, and communication skills for managing multiple parallel workstreams.
- Ability to critically review code and experiment results, including work produced by AI agents.
- Sincere interest in developing safe and beneficial AI systems.
- Preferred experience leading ambiguous technical projects and setting direction for engineers or researchers.
- Preferred experience evaluating large language models, developing classifiers, or operating high-scale production systems.
- Preferred experience building rare-event detection systems such as abuse, fraud, spam, or intrusion detection.
- Preferred experience operating low-latency, high-availability services, including on-call responsibilities.
- Preferred experience fine-tuning language models, using probes, or reading model internals.
- Preferred experience with LLM inference or serving infrastructure and training infrastructure integration.
- Preferred background in security, sandboxing, detection and response, AI control, scalable oversight, agent safety, or reinforcement learning.
Benefits
- Annual salary of £1—£1 GBP.
- Visa sponsorship is offered, with reasonable efforts made after an offer and support from an immigration lawyer.
- Hybrid work policy requiring staff to be in an office at least 25% of the time, with some roles requiring more office time.
- Competitive compensation and benefits, optional equity donation matching, generous vacation and parental leave, flexible working hours, and office collaboration space.
Categories
About Anthropic
Anthropic builds large language models and the Claude AI assistant for developers and enterprises, offered via API access and enterprise plans. Founded in 2021 and headquartered in San Francisco, it distributes Claude through its own platform and via partners such as Amazon Bedrock and Google Cloud’s Vertex AI. Its work emphasizes model reliability, interpretability, and practical tooling for tasks like coding assistance, analysis, and customer support automation.
