HD Hyundai Robotics cuts problem-solving time 63% with AI agents

HD Hyundai Robotics presented results from deploying AI agents in manufacturing operations at Palantir's Sovereignty Bootcamp in New York. Since August 2026, the company reduced average problem-solving time from 21.5 days to 7.9 days using AI agents integrated with Palantir's platform. The system replaces manual trial-and-error robot programming with AI-generated optimization suggestions that engineers validate against field experience.

Key points

• 63% faster turnaround: Field problem resolution dropped from 21.5 days to 7.9 days on average after introducing AI agents in August
• 50%+ cost reduction target: AI agents analyze existing robot programs and suggest improvements to movement paths, task sequences, and speeds, aiming to cut programming costs in half
• Role transformation: Engineers shift from manual robot programming and commissioning to defining problems and implementing AI-driven solutions as "field solution experts"
• Palantir partnership: Regular steering committee meetings set priorities; future plans include connecting HD Hyundai robots directly to Palantir's AIP platform

How to use it

• Replace iterative tuning: Instead of engineers manually adjusting robot movement, speed, and task order on-site, have AI agents parse existing programs and propose optimizations for engineer review
• Standardize tribal knowledge: Use AI to capture and apply the expertise that previously resided only with seasoned engineers, making robot deployment more repeatable
• Combine simulation + AI: Run AI-suggested robot configurations through simulation before field deployment to validate improvements without production downtime

Has anyone tested AI agents for industrial automation workflows? What time or cost savings did you see?