Applied Research - Forward-Deployed RL

Prime Intellect AI; San Francisco (CA); Remoto; USD 150.000 - 300.000; Tempo pieno

Vantaggi offerti da questo lavoro

Visa sponsorship

Relocation assistance

Professional development budget

Team off-sites & conference attendance

Descrizione del lavoro

Prime Intellect AI seeks a role focused on applying RL programs from problem framing to deployed improvements. You will design environments, evaluations, and baselines, run post-training experiments, and deploy results that become part of Prime's shared stack.

You will work directly with customer researchers, engineers, and leaders, taking accountability for both research outcomes and practical impact. The role emphasizes external-facing work, with opportunities across technology, science,

Competenze

• Experience in customer-embedded technical work or applied ML projects.
• Ability to connect experimental evidence to business decisions without lowering research quality.
• Hands-on experience building and debugging model-training, post-training, or agent loops.
• Strong Python and the ability to create reusable artifacts for researchers and engineers.
• Comfort handling ambiguous problems and keeping partners updated as you resolve them.

Mansioni

• Own applied RL programs from the first customer problem to deployed improvements.
• Design environments, baselines, evaluations, and measurement plans.
• Run post-training experiments and deploy results that generalize into Prime's stack.
• Collaborate with customer researchers, engineers, and leaders; accountable for both results and impact.

Conoscenze

Experimental judgment

Stakeholder communication

Research engineering

Post-training ML

Strumenti

Python

RL training

Environment engineering

Descrizione del lavoro

Own Your Intelligence

Prime Intellect is building the open superintelligence stack: the infrastructure frontier AI labs build internally, made available to every ambitious AI team.

Our platform, Lab, unifies compute, environments, evaluations, secure sandboxes, high-performance training, and deployment into one full-stack system for post-training at frontier scale - from SFT and RL to tool use, agent workflows, and continuously improving production models. We are building open frontier AI: open-source models trained end to end for long-horizon tasks like autonomous research, and the full-stack platform our own research team uses to build them. The next generation of AI companies, enterprises, and research teams do not just need more GPUs. They need the ability to turn their own workflows, tools, data, and feedback loops into superintelligence they own.

Prime Intellect has raised $150M in total funding from Founders Fund, Radical Ventures, NVIDIA, and exceptional AI, infrastructure, and enterprise operators — including Andrej Karpathy, Dwarkesh Patel, and leaders and founders from Ramp, Perplexity, Harvey, Mercor, Zapier, Datadog, Cognition, OpenAI, Thinking Machines, Together AI, SemiAnalysis, LangChain, Browserbase, Cloudflare, Sierra, Databricks, Airbnb, OpenRouter, Standard Intelligence, Fleet, Core Auto, and more. We are looking for people who want to build at the intersection of frontier research, real infrastructure, and go-to-market for a category that does not fully exist yet.

Role Impact

You will own applied RL programs from the first ambiguous customer problem through a deployed improvement loop. You will understand the workflow and business stakes, decide what can be measured and learned, build the environment and evaluation, run post-training experiments, deploy the result, and determine what should generalize into Prime's shared stack. The majority of your work will be organized around external programs. You will work directly with customer researchers, engineers, and leaders, and you will be accountable for both the research result and its practical impact.

Responsibilities

Applied Research at Prime

We turn a workflow into tasks, environments, evaluations, and reward signals, run post-training experiments against real baselines, and deploy what works. When a result looks good, we find out whether the gain came from the model or from the data, harness, inference, or infrastructure. The reusable parts become shared research, open-source systems, and product capabilities.

Every Applied Research role is externally facing. Most researchers work directly with customers, users, or research partners while maintaining a substantive research agenda.

Role scope

We have active programs across technology, science, engineering, regulated industries, and the public sector.

Representative early problems

• Turn a partially documented workflow and business goal into an environment, baseline, and evaluation that the customer team agrees measures the right outcome.

• Improve a domain agent through RL or another post-training method while separating model limitations from task, harness, inference, and integration failures.

• Carry a successful experiment into deployment, measure its effect on the customer outcome, and convert the recurring parts into reusable environments, recipes, data loops, or platform capabilities.

What you will own

• Technical and business discovery that produces a clear research plan, success measure, environment, and baseline.

• RL, distillation, and other post-training experiments tied to deployment outcomes.

• Agent systems that handle tool use, memory, recovery, long horizons, and human oversight.

• Deployment and adoption decisions, plus reusable artifacts carried back to Prime's research, engineering, and product teams.

Requirements

• Relevant experience: customer-embedded technical work, research engineering, technical founding experience, applied ML consulting with real build ownership, or end-to-end post-training programs.

• You should be comfortable with senior external teams and able to connect 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: meaningful baselines, careful trace inspection, clear failure attribution, and a willingness to change direction when the evidence changes.

• Strong Python and the 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 you resolve them.

Evidence of exceptional work

We value publications, open-source systems, benchmarks, and shipped products as evidence of exceptional work. Show us something difficult you built, what failed, and how the evidence changed your approach.

Benefits and Perks

• Cash Compensation Range of $150-300k + equity incentives

• Flexible Work (San Francisco, New York, London, or remote)

• Visa Sponsorship & relocation support

• Professional Development budget

• Team Off-sites & conference attendance