Applied Research - Forward-Deployed RL

Prime Intellect•San Francisco, CA

1d•$150,000 - $300,000

About The Position

Prime Intellect is hiring an Applied Researcher to own applied reinforcement learning programs, primarily organized around external customer programs. The role spans the full process from an ambiguous customer problem through a deployed improvement loop: understanding workflows and business stakes, defining measurable outcomes, building environments and evaluations, running post-training experiments, deploying results, and deciding what should be incorporated into Prime’s shared stack. The researcher will work directly with customer researchers, engineers, and leaders, and will be accountable for both research quality and practical impact. Programs span technology, science, engineering, regulated industries, and the public sector. The work includes translating workflows into tasks, environments, evaluations, and reward signals; testing against real baselines; diagnosing whether observed gains come from the model or from data, harnesses, inference, or infrastructure; and turning reusable results into shared research, open-source systems, and product capabilities.

Requirements

Relevant experience in customer-embedded technical work, research engineering, technical founding, applied ML consulting with real build ownership, or end-to-end post-training programs.
Comfort working with senior external teams and connecting experimental evidence to business decisions without lowering the research bar.
Hands-on experience building and debugging model-training, post-training, agent, evaluation, or closely adjacent experimental loops.
Strong experimental judgment, including meaningful baselines, careful trace inspection, clear failure attribution, and willingness to change direction when evidence changes.
Strong Python skills and ability to build artifacts that other researchers and engineers can run, inspect, and extend.
Comfort owning ambiguous problems and keeping partners and teammates current as problems are resolved.

Nice To Haves

Publications, open-source systems, benchmarks, or shipped products demonstrating exceptional work.
Ability to show a difficult project built, what failed, and how evidence changed the approach.

Responsibilities

Conduct technical and business discovery and produce a clear research plan, success measure, environment, and baseline.
Build tasks, environments, evaluations, and reward signals from customer workflows and goals.
Run RL, distillation, and other post-training experiments tied to deployment outcomes.
Build and improve agent systems that handle tool use, memory, recovery, long horizons, and human oversight.
Inspect experimental evidence and attribute gains or failures to the model, data, harness, inference, or infrastructure.
Carry successful experiments into deployment, measure their effect on customer outcomes, and make deployment and adoption decisions.
Convert recurring work into reusable environments, recipes, data loops, or platform capabilities for Prime’s research, engineering, and product teams.
Work directly with customer researchers, engineers, leaders, users, or research partners while maintaining a substantive research agenda.
Keep partners and teammates current while resolving ambiguous problems.