Applied Research Engineer, Agents

About this role

Labelbox seeks an Applied Research Engineer to advance AI agents by designing data pipelines, benchmarks, and training frameworks for autonomous systems. You'll collaborate with frontier AI labs on cutting-edge agent research, turning prototypes into scalable products while publishing findings at top venues.

What you'll do

• Create frameworks and tools to construct, train, benchmark, and evaluate autonomous agent capabilities
• Design agent-focused data programs using supervised fine-tuning (SFT) and reinforcement learning (RL)
• Develop data pipelines from diverse sources including code repositories, web browsers, and computer systems
• Engage with frontier AI lab customers to understand evolving agent data requirements
• Implement and adapt open-source agent libraries and benchmarks with proprietary datasets
• Publish research findings in academic journals, conferences, and blog posts

What they're looking for

• LLM agents and planning/execution loops
• Supervised fine-tuning (SFT) and reinforcement learning (RL) methodologies
• Python and deep learning frameworks (PyTorch, JAX, TensorFlow)
• Agent benchmarking and evaluation
• Research paper writing and publication
• Data pipeline development
• Problem-solving in ambiguous situations
• Proficiency with frontier model architectures and post-training techniques

Benefits

• Work on cutting-edge AI research at the frontier of agent development
• Collaborate with leading research labs and industry experts
• High-impact environment with rapid career growth tied to contributions
• Clear ownership and autonomy to drive results with measurable impact
• Continuous learning opportunities in advanced AI and data-centric machine learning