Applied AI Researcher — Frontier Agent Systems (Remote)

Prime Intellect, Inc.; Oregon; Hybrid; USD 150,000 - 300,000; Full time

Benefits offered by this job

Cash compensation range $150k-$300k +

Equity incentives

Flexible work (SF, NY, London orRemote

Visa sponsorship & relocation support

Professional development budget

Team off-sites & conference attendance

Job summary

Prime Intellect, Inc. is seeking an Applied Researcher to build post-training and evaluation systems for multimodal agents operating across browsers, desktops, and cross-application workflows.

You will work on environments, rewards, and experiments to drive reliable improvement, often in collaboration with customers and research partners. You will own the development of environments from real workflows and contribute to open-source systems and platform capabilities, while maintaining a strong

Qualifications

• Experience with multimodal agents, browser or desktop automation, RL or imitation learning.
• Hands-on in model-training, post-training, evaluation loops.
• Strong experimental judgment and traceability of results.
• Proficient Python and ability to build reproducible artifacts.

Responsibilities

• Turn workflows into tasks, environments, evaluations, and reward signals.
• Run post-training experiments against real baselines and deploy what works.
• Collaborate with customers, users, or research partners while pursuing a research agenda.

Skills

Multimodal agents

Browser/Desktop automation

RL / imitation learning

HCI / long-horizon evaluation

Python proficiency

Experimentation judgment

Tools

Python

Experimentation frameworks

Job description

Prime Intellect, Inc. is seeking an Applied Researcher to build post-training and evaluation systems for multimodal agents operating across browsers, desktops, and cross-application workflows.

You will work on environments, rewards, and experiments to drive reliable improvement, often in collaboration with customers and research partners. You will own the development of environments from real workflows and contribute to open-source systems and platform capabilities, while maintaining a strong