Staff AI Engineer - Post-Training & Agentic RL

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

Job summary

Hark in San Jose, California is seeking a Member of Technical Staff to lead the development of strategies for coding agents using reinforcement learning. The role involves designing experiments, building evaluation frameworks, and collaborating across teams to implement improvements driven by research insights.

Strong candidates should have a background in machine learning, particularly within reinforcement learning, with skills in Python and PyTorch, and an ability to work in fast-paced, research-oriented settings.

Qualifications

• Strong background in machine learning with experience in large model training.
• Deep understanding of reinforcement learning concepts and techniques.
• Proficiency in Python and experience with PyTorch.
• Ability to design rigorous experiments and diagnose issues.

Responsibilities

• Design and implement post-training strategies for coding agents.
• Build simulation and scaffolding environments for agentic RL.
• Develop reward modeling pipelines for training agents.
• Scale synthetic data generation and improve training efficiency.
• Build evaluation frameworks to measure the progress of agents.
• Collaborate to translate research insights into improvements.

Skills

Machine learning

Reinforcement learning

Python

PyTorch

Simulation environments

Job description

Hark in San Jose, California is seeking a Member of Technical Staff to lead the development of strategies for coding agents using reinforcement learning. The role involves designing experiments, building evaluation frameworks, and collaborating across teams to implement improvements driven by research insights.

Strong candidates should have a background in machine learning, particularly within reinforcement learning, with skills in Python and PyTorch, and an ability to work in fast-paced, research-oriented settings.