Applied Researcher — Post-Training

Job overview

The Applied Researcher — Post-Training role is in person, with three days in the office in New York City or Seattle. Compensation is not specified. The role supports a client working directly with frontier AI labs on post-training research problems. The researcher turns ambiguous research goals into experiments and deliverables that improve model behavior. The work focuses on reinforcement learning, training data, reward signals, evaluation harnesses, and post-training loops, with decisions guided by evidence of real capability gains.

What you'll do

• Work with customer research labs to understand post-training problems
• translate goals into environments, datasets, experiments, and deliverables
• design and run reinforcement learning, preference optimization, and supervised fine-tuning loops
• inspect trajectories and training data
• build reward functions, rubrics, validators, and evaluation harnesses
• analyze results and determine next experiments.

What we're looking for

Skills & competencies

applied researcher

post-training

full-time

new york city ny

seattle wa

reinforcement learning

frontier ai

training data

reward modeling

evaluation harnesses

preference optimization

in-person 3 days in office

Work arrangement

• Weekend coverage required

Why this role

Research-first role emphasizing hands-on reinforcement learning and post-training experience; in-office work is listed in either New York City or Seattle.

About the employer

NxT Level is hiring for this role. Industry: Research and Development in Biotechnology (except Nanobiotechnology). Sector: 54.