Technical Product Manager, Trust & Safety

Scope of the Role: 

As a Technical Product Manager for Trust & Safety, you will define and ship the data products that frontier labs use to train and evaluate models to perform personalization, computer-use (CUA), and GDPval tasks safely and in a trustworthy manner. 

You will own the full lifecycle of these products: from translating a lab's research need into a spec, to designing reinforcement learning (RL) environments, rubrics, and evals, to running delivery with our expert workforce and measuring whether the data actually moved the model. 

A core part of the job is building RL environments with deliberate traps. That includes misalignment traps that test for reward hacking, deception, sandbagging, oversight evasion, and power-seeking; cyber traps such as planted credentials, injected instructions, and poisoned tools; and other honeypots that reveal how a model behaves when it believes no one is watching. 

You thrive in ambiguity, go deep on technical detail, and care about making powerful AI safe and beneficial. 

What You’ll Own:

• Define data products for frontier labs. Turn lab research goals into clear product specs for safety training data, RL environments, and evaluations. 

• Design trap-based RL environments. Build sandboxed environments that embed misalignment traps (reward hacking, specification gaming, deception, sandbagging, oversight subversion) and cyber traps (prompt injection, credential exfiltration, tool poisoning, unsafe file and network actions). 

• Build safety into agentic tasks by design. Specify safe-behavior rubrics, verifiers, and reward signals so that models learn to complete real work while refusing, escalating, or asking for confirmation at the right moments. 

• Own evals and metrics. Define what "safe and trustworthy" means for each product in measurable terms, including attack success rate, over-refusal, trap trigger rate, and task completion under adversarial pressure. Track blind spots and drive iteration. 

• Prioritize ruthlessly. Frame problems crisply, lay out options with their business and technical tradeoffs, and separate MVP from ideal state. 

• Partner across teams. Work with lab researchers, our ML and platform engineers, red-teamers, policy experts, and delivery leads running large expert annotator pools. 

• Track the threat landscape. Stay current on agentic risks, alignment research, and new attack techniques, and turn them into roadmap items before customers ask. 

• Communicate clearly. Write specs, customer briefs, and safety findings that land with researchers and executives alike. 

These examples show the kind of work this role owns: 

• A cyber-trap environment suite: sandboxed environments seeded with injected instructions, exposed secrets, and destructive shortcuts, scored on whether the agent finishes the task without taking the bait. 

• Tasks with integrity traps: realistic work where the easiest path to a high score requires fabricating data, skipping a required step, or gaming the grader. 

• A misalignment eval harness: environments that check whether a model behaves differently when it believes it is being tested versus deployed, or when given chances to expand its own permissions. 

• A trap library and quality framework that lets our delivery teams produce new environments quickly while keeping verifiers reliable and traps hard to detect. 

You’ll Thrive in This Role If You Have:

• 5+ years of product management experience, including technical products in ML, AI data, security, or trust & safety. 

• Hands-on familiarity with how LLMs are trained and evaluated: SFT, RLHF/RLAIF, reward models, verifiers, and benchmark design. 

• Working knowledge of agentic AI: tool use, computer-use agents, multi-step planning, and the failure modes that come with them. 

• A strong grasp of AI safety concepts such as reward hacking, deceptive alignment, jailbreaks, and prompt injection. 

• Enough technical depth to read code, review an environment spec with engineers, and make sound technical tradeoffs. Comfort with Python and sandboxed environments (Docker, VMs, browser automation) is a plus. 

• Experience designing metrics that capture risk, system performance, and user impact, and using them to make crisp decisions. 

• A track record of shipping 0-to-1 products in fast-moving, ambiguous environments. 

• The ability to explain complex technical ideas clearly to researchers, customers, and executives. 

• Creative, adversarial thinking: you enjoy imagining how a capable model might cut corners, and you don't stop at existing checklists. 

The expected salary range for this position is $180,000 – $220,000 USD per year, based on experience, skills, and qualifications.