Find your next role
Discover roles across ambitious teams
18,714 jobs
- More jobs
Toptal
ML Inference Engineer, Terminal-Bench Task Designer
Remote · Mid level
About the role
Design expert-level Terminal-Bench tasks that simulate real ML inference work for AI agents to solve. Build reproducible environments, reference solutions, and automated checks for task validation.
What you’ll do
• Design expert-level Terminal-Bench tasks from real ML inference work. • Define files, systems, and environments for realistic challenges. • Write and verify working reference solutions. • Build automated checks for result correctness. • Package tasks with Docker and Git, document systems and rules.
What you’ll bring
• 3+ years hands-on ML inference or PhD with active research. • Experience with real scale model serving (multi-GPU or production). • Hands-on with PyTorch or JAX internals and serving stacks. • Technical fluency in Python, command line, files, APIs. • Availability to start early November 2026.