RobotWorld

Building Robots That Actually Help People: Lessons from Charlie Kemp and the Stretch Robot

By RobotWorld·9/30/2026

What does it take to build a robot that genuinely improves someone's life? For Charlie Kemp — cofounder and CTO of Hello Robot, and a professor at Georgia Tech — the answer starts not in the lab, but with the person sitting across from you.

Kemp's career arc is a useful lens for understanding where assistive robotics is heading, and why so much of the field's progress depends on listening as much as engineering.

From AI Research to Real-World Impact

Kemp's path followed a familiar route through elite academic institutions, including AI research at MIT, before landing in a space that blends robotics, human factors, and disability studies in ways that few engineering programs explicitly prepare you for.

That pivot — from theoretical AI to practical assistive hardware — is more significant than it might appear. It reflects a broader shift in robotics thinking: the field has moved away from asking "what can a robot do?" toward asking "what does a specific person actually need a robot to do?" Those are very different design questions, and they lead to very different machines.

What Makes Stretch Different

Hello Robot's flagship product, Stretch, is a tall, slender mobile manipulator designed to assist people with limited mobility in their homes. It can reach objects on shelves, retrieve items from the floor, open drawers, and help with dozens of everyday tasks that many people take for granted.

Stretch deliberately avoids the imposing, industrial look of many research robots. It's lightweight, relatively affordable by research standards, and designed to operate safely in close proximity to humans — an environment far messier and less predictable than any factory floor.

That design philosophy comes directly from Kemp's years of collaboration with people with disabilities. Working alongside users who need genuine help — not a proof-of-concept demo — forces an engineer to confront every assumption baked into their original design. Does the robot's approach speed feel threatening? Can someone with limited hand function actually command it? Is the interface understandable when you're fatigued?

These aren't software bugs. They're fundamental design questions that only emerge through sustained, honest engagement with the people a robot is meant to serve.

The Human-Centered Design Principle

Kemp's approach is a textbook example of what engineers and UX researchers call human-centered design — but applied in a domain where the stakes are unusually high. For someone with a spinal cord injury or a progressive neurological condition, a robot that works most of the time isn't good enough. Reliability, predictability, and ease of use aren't nice-to-have features; they're the entire product.

This has practical implications for how robots are tested and iterated. Rather than evaluating a robot purely against technical benchmarks — can it pick up object X with Y% success rate? — Kemp's team measures outcomes in terms of user autonomy and quality of life. Those metrics are harder to quantify, but they're the ones that matter.

It also shapes what gets built in the first place. Many robotics projects start with a capable technology and then search for applications. Kemp's approach inverts that: start with a real, documented human need and work backward to the technology required to meet it.

Advice for Aspiring Roboticists

Kemp's advice for those entering the field is grounded in that same perspective. He emphasizes the importance of working directly with end users early in the development process — not as a final validation step, but as a core part of research. He also stresses the value of interdisciplinary thinking: robotics increasingly draws on fields like occupational therapy, cognitive science, and participatory design, not just mechanical and software engineering.

For students and early-career engineers, this is a meaningful nudge toward breadth. Understanding how a gripper mechanism works matters, but so does understanding why someone might refuse to use a robot that enters their bedroom uninvited. Both are engineering problems, just at different levels of abstraction.

Why This Matters for the Broader Field

Assistive robotics sits at an intersection of aging demographics, healthcare costs, and rapid hardware progress. As robotic platforms become more capable and more affordable, the limiting factor for real-world deployment is increasingly not the technology itself — it's the human factors work required to make that technology genuinely usable.

Kemp's career illustrates that this work is not a footnote to robot development. It is robot development. The engineers, researchers, and students who internalize that lesson early will be the ones building machines that people actually want to live and work alongside.

For those looking to develop foundational robotics skills — from programming manipulator arms to understanding how mobile platforms navigate real environments — hands-on platforms like the Hiwonder TonyPi Pro for humanoid motion research or the NVIDIA Jetson Orin Nano Super Developer Kit for on-device AI inference provide practical starting points for exploring the same technical challenges that underpin systems like Stretch.

The path from a classroom robot to an assistive device in someone's home is long — but it starts with exactly the kind of curiosity, empathy, and technical ambition that Kemp's story models so well.

Interested in exploring robotics platforms for education, research, or assistive applications? Browse our range of hands-on robotics kits and AI hardware, or get in touch with our team for guidance on the right platform for your project.

References

This article was drafted with AI assistance and reviewed before publishing.

Related reading