🛠 Tech Stack

💼 About This Role

You'll own compute planning, scientific RL environment health, and vendor pipelines for Anthropic's Discovery team, which trains large-scale models to build an AI scientist. You'll drive the pace of research by coordinating across research, infrastructure, and data operations. This role combines deep technical debugging skills with organizational effectiveness in a fast-moving AI research setting.

🎯 What You'll Do

• Manage compute planning across SL, RL, and sandboxing workloads
• Monitor scientific RL environment health and drive issue resolution
• Expedite external vendor pipeline for RL environments
• Collaborate with research teams to translate goals into roadmaps
• Establish processes to bring structure to research without slowing it

📋 Requirements

• ML engineering or STEM R&D background with transition to program management
• Hands-on experience with ML training pipelines and RLHF systems
• Track record of building execution plans and high-leverage processes
• Ability to navigate ambiguity and fast-growing organization

✨ Nice to Have

• Experience in life sciences or bio R&D
• Knowledge of reward design and quality review in RL
• Past work in AI safety or interpretability

🎁 Benefits & Perks

• 💰 Annual salary range $365k-$435k USD
• 🏢 Hybrid policy with at least 25% in-office (San Francisco or New York)
• 🌍 Visa sponsorship available
• 🤝 Opportunity to work on cutting-edge AI safety

📨 Hiring Process

Estimated timeline: 2-4 weeks · AI estimate

1. 1Recruiter Phone Screen· 30 min
2. 2Hiring Manager Interview· 45 min
3. 3Technical Deep Dive· 60 min

🚩 Heads Up

• Role expects both deep ML engineering expertise and program management skills
• Hybrid policy (at least 25% in-office) may not suit fully remote candidates
• Visa sponsorship not guaranteed for all roles