Shift AI Podcast How Robots are Learning to Work Like Humans with Acumino CEO Patrick Jarvis
Aug 4, 2025
In this engaging discussion, Patrick Jarvis, President and Chief Innovation Officer at Acumino, shares his extensive experience in AI and robotics. He unveils groundbreaking advancements in robotic reliability and dexterity for tasks like fine assembly. Jarvis emphasizes how robots can empower humans by simplifying AI training processes. The conversation explores the future of robots in both industrial and home settings, highlighting their potential to improve productivity and quality of life. Tune in for insights on the evolving relationship between technology and work!
AI Snips
Chapters
Transcript
Episode notes
Patrick Jarvis' Robotics Journey
- Patrick Jarvis shared his journey working with trillion-dollar companies on AI problems for nearly 20 years.
- He founded Acumino to solve ultra-reliable robotic manipulation, focusing on robots assembling engines and threading screws like humans.
Robots Achieve Fine-Tuned Touch
- Robots now can do fine manipulation tasks like assembling engines or threading screws, tasks once considered almost impossible.
- Success requires robots to sense touch and force accurately to handle part variations in real time.
The Unlock Behind Reliable Robots
- Advances in large language models and vision by companies like NVIDIA and Google unlocked new robot capabilities.
- However, the critical missing piece was high-fidelity force and torque data to enable physical interaction with objects.