Senior Frontier RL Environments Engineer

Hark; San Jose (CA); On-site; USD 180,000 - 450,000; Full time

Job summary

Hark is seeking a Member of Technical Staff, Frontier RL Environments to build agent environments for proactive, personal intelligence. You will design RL environments and curricula, scale infrastructure, and collaborate with research, engineering, product, and safety teams to advance training runs.

This role sits at the frontier of environment design, evaluation science, and large-scale model training, requiring hands-on engineering and strong collaboration across teams to push multimodal

Qualifications

• Strong background in machine learning with hands-on experience training or fine-tuning large models (LLMs, multimodal, or equivalent systems).
• Hands-on experience building RL environments, simulators, or task suites (Gym-style interfaces, game engines, browser/OS automation sandboxes, robotics/hardware simulators).
• Direct work with LLMs and post-training techniques: RL, RLHF/RLAIF, reward modeling and graders, synthetic data pipelines, or agentic/tool-using systems.
• Track record on fuzzy problems with noisy data and ambiguous objectives requiring experimentation.
• Strong point of view on making a personal assistant useful, trustworthy, and pleasant to rely on daily.

Responsibilities

• Design and build RL environments and tasks spanning long-horizon planning and multimodal interaction.
• Architect reward functions, graders, and curricula for environments.
• Scale environment infrastructure for thousands of tasks across simulations and real sessions.
• Shape decisions on large training runs and assess intelligence capabilities.
• Create self-improvement loops where models generate and refine training environments.

Skills

ML fundamentals

RL environments

LLMs / multimodal

Python / tooling

Cross-team collaboration

Education

Advanced degree in ML or CS

Tools

Gym-style interfaces

Simulation tools

Robotics simulators

Job description

Hark is seeking a Member of Technical Staff, Frontier RL Environments to build agent environments for proactive, personal intelligence. You will design RL environments and curricula, scale infrastructure, and collaborate with research, engineering, product, and safety teams to advance training runs.

This role sits at the frontier of environment design, evaluation science, and large-scale model training, requiring hands-on engineering and strong collaboration across teams to push multimodal