Research Engineer, Agentic Systems

About this role

Mirage seeks an ML engineer to develop agentic systems that leverage large language models for multimodal creative tasks, particularly video editing and analysis. You'll design end-to-end agent pipelines, advance LLM reasoning capabilities, and create evaluation frameworks for real-world creative workflows.

What you'll do

• Design and build end-to-end agentic systems for creative video tasks
• Develop novel training and adaptation approaches for LLMs powering agents
• Create objectives, datasets, and fine-tuning strategies to improve agent behavior
• Explore multimodal reasoning and structured generation for creative control
• Run systematic experiments to evaluate and improve agent performance
• Analyze failure modes across the full agent loop and iterate on improvements

What they're looking for

• Large language models and transformer architectures
• Agentic system design and tool-use implementation
• Fine-tuning, alignment, and post-training methods
• Multimodal machine learning and structured generation
• Experimental design and evaluation frameworks
• Production ML systems and full-stack ownership
• Video understanding and temporal reasoning
• Python and ML engineering best practices

Benefits

• Comprehensive medical, dental, and vision plans
• 401(k) with employer match
• Catered lunch multiple days per week and dinner stipend for late work
• Commuter and Grubhub benefits
• Multiple team offsites annually with monthly team events
• Generous PTO policy