FIRST SEENJUN 21, 2026RUNREQ OBSERVATIONDAYS OPEN9555D COMPANY MEDIANTEAM OPEN66AI RESEARCH & ENGINEERING30D TEAM+7NEWLY OBSERVEDTRACKED COMP$350K–$850K /YRDISCLOSED RANGE / ALL TIERSSOURCECLOSEDCHECKED 20D AGO
JOB DESCRIPTIONSHORTENED FROM SOURCE
Key responsibilities
4 ITEMS
Own data strategy for knowledge work, from task sourcing through reinforcement learning training.
Create reinforcement learning environments by identifying valuable tasks, designing rewards, managing vendors, and measuring model impact.
Build data pipelines, evaluations, and quality assurance frameworks that detect reward hacking and measure generalization.
Partner with domain experts, research teams, and product teams to turn capability goals into training environments and evaluations.
Qualifications needed
4 ITEMS
Bachelor’s degree in a relevant field, or equivalent education, training, or experience.
Experience fine-tuning large language models for specific domains or real-world uses.
Experience with reinforcement learning, reward design, or training data curation for large language models.
Preferred: Experience training production machine learning systems, designing large language model evaluations, or working with technical vendors.
EVENT>Anthropic :: Research Engineer, Domain Scaling first observed 95 days ago