Norbert Health

Every nurse learned their job from a preceptor: an experienced nurse who trained them on the floor, patient by patient, until they were competent to practice. Not in a classroom. At the bedside. 🩺 Robots entering healthcare will need the same thing. Humanoid robots will be in our homes. 2 years? 10? Nobody knows exactly. But they're coming. And hardware alone won't make them useful in care. A robot becomes competent the way a nurse does: by acquiring skills. ✅ Measuring a dangerously high pulse, retaking it twice to be sure, then driving to the nursing station to find the nurse because a dashboard alert isn't enough. That's a skill. Live today on our floors. ✅ Detecting a UTI days before symptoms appear, from subtle shifts in gait, confusion, and daily routine. That's a skill our robots are learning now. ✅ Running a cognitive screen disguised as a friendly conversation, so the resident never feels tested. A skill. ✅ Predicting a fall before it happens, and getting PT to the resident proactively. A skill. Each one has to be built, then learned on real care floors with real patients. The sensing takes years of medical-grade engineering and regulatory work. The conversation and pattern skills take something you cannot compress: months of daily encounters and longitudinal data from thousands of real interactions. This is what Norbert Health is: the preceptor for physical AI in healthcare. 🤖 Our robots are live in skilled nursing facilities today, monitoring hundreds of patients daily, adding skills one by one. And every skill is decoupled from any single robot body, so they transfer to every platform that comes next. The robots that will care for us at home in 2030 are learning their skills on care floors in 2026. Hardware makes a robot capable. Skills make it competent. That's the layer we build. 💙

Meet Jinsy J., our new Chief Medical Officer ✨ A physician who's spent her career bridging clinical medicine and healthcare innovation, from inside health systems and as an investor. At Norbert, she'll sharpen the clinical thinking behind everything our robots do: what we measure, how we screen, and how care teams actually use what we surface. Having a clinical mind at the core of product decisions changes what we're able to build. Welcome, Jinsy.🤗

Last week, a patient about to be discharged said goodbye to the robot in her room: "I'm going home tomorrow. I'll miss you. Norbert" Norbert congratulated her on the good news. Moments like this are happening more and more often, and we think we know why. Last month, our avatar, the blue teddy bear, got new skills. Nonverbal communication and body language are essential components of human communication, so we decided to enable our avatar to move and emulate this nonverbal communication. It waves hello, nods along while patients talk, and gives a thumbs-up when they share good news. Also, it knows when you focus on Norbert or wave at it. And this works.✨ Before the launch, 34% of patients engaged with Norbert during visits: they talked to him, or showed they knew he was there. Since the launch, engagement has risen to 48%. Patients who never used to respond now wave back and smile. Some can't stop waving - and Norbert can’t stop waving back. When one patient's grandchildren came to visit, they waved at Norbert, too, and the whole family gathered around to record videos of Norbert. Rooms filled with joy. 🐻🤳 Behind the warmth, this matters clinically. When a patient chats with Norbert every day, changes in mobility, alertness, and biomarkers surface the early signs of decline that are easiest to miss between nurse visits. A wave back is a data point. A daily hello is a daily check-in. 👋 To every patient who helped design Norbert's new face: thank you. 🩵 What should Norbert learn next? #HealthcareRobotics #PhysicalAI #HumanRobotInteraction #RoboticCare #HealthTech #AgeTech #NorbertHealth

What does it take for a robot to earn a place in healthcare? 🤖🏥 The hard part was never just the robotics. It’s building something patients trust, clinicians actually use, and care facilities can run every day. At Figma #Config2026, our Head of Product & Design, Anna Oh, shared what years of deploying robots alongside patients and clinical teams have taught us about healthcare robotics, physical AI, and human trust. The response has been incredible. Product builders, designers, and researchers have shared how the talk changed the way they think about aging, responsible robotics, and the future of AI in healthcare. Successful healthcare robotics is defined by more than what a robot can do. It depends on whether people accept it, trust it, and choose to build a relationship with it. The full talk is finally live on YouTube.  🎥 Watch it here: https://lnkd.in/g3pvYUru Let us know what you think ✨ #HealthcareRobotics  #PhysicalAI #HumanRobotInteraction #RoboticCare #HealthTech #NorbertHealth

We just visited one of the most impressive rehab facilities we've ever seen. Stunning space. Top-tier staff. Consistently full census. They want Norbert for its ability to monitor patients' vital signs, behavior, and mobility habits. The clinical case is straightforward: a facility this size needs dense medical and behavioral monitoring to catch early signs of decline before they become larger issues such as hospital readmissions. Integrating physiological, behavioral, mobility, and other indicators to improve the performance of their care programs every day. That's the core of what we built Norbert to do. But as we walked the floor, a second use case kept coming up. The facility is large. Getting from one end to the other takes time. And a meaningful part of a nursing assistant's day disappears into logistics and responding to frequent requests from residents: answering questions about the care plan and the day's schedule, addressing small needs, administering standard questionnaires, asking follow-up questions, and relaying reminders. Important to patients, and time to process queries for residents is one of the facility's quality metrics. But this is not where great caregivers should spend their hours each day. Norbert could be rounding those rooms instead and chatting with residents. Answering questions, taking requests, following up reliably, every time, without forgetting. And while it does, it quietly listens for changes in mood or affect that signal a resident needs human attention. The result they're expecting: quality of service metrics up immediately. And nursing assistants are free to do what no robot will ever do, as well as they can: being present, being human, delivering informed care. The robot roams the halls with its notepad. The nurses care and understand humans. This is what Norbert is built for: giving caregivers back the time and energy to be fully present for every patient who needs them, which directly improves both patient outcomes and experience, and staff burnout. If you run a rehab facility, a skilled nursing facility, or a senior care community and want to see what this looks like in practice, we’d love to show you. Contact us today! #HealthcareRobotics #RoboticCare #HealthTech #FutureOfCare #NorbertHealth

Our CEO Alexandre Winter shares why healthcare robotics deserves its own category, and why solving care challenges requires a different approach than traditional robotics.🤖🏥

Norbert Health is now a graduate of the Endless Frontier Labs 2025–26 cohort, Digital Health track. 🏁 We are one of 74 ventures across Deep Tech, Digital Health, Digital Tech, and Life Sciences. 🏆 NYU-EFL has been truly impactful for us: 🤝 We met spot-on potential partners we will integrate in our platform 💵 We were connected with investors who invested, or will soon invest, our latest financing round 🦾 We were connected with potential customers that are moving fast and will likely close soon 💡 We received deep and insightful, often impactful feedback from a set of very sharp mentors A huge thank you to: 🙏 our mentors - special mention to Jinsy J., Manuel Jaime, Andy Feldstein, DSc, Diegane Dione 🎓 the NYU Stern School of Business MBA team, 🎯 our cohort-mates 👨🏫 Pr. Deepak Hegde and the Endless Frontier Labs team for building a program that actually does what it says. And helps us grow! EFL is among the world’s best accelerators, takes no equity, charges no fees, and is open to founders anywhere in the world. Applications for the 2026–27 cohort are open: https://lnkd.in/gxTU4Jds We highly recommend it! Feel free to reach out to us to get more info or intro to the program.  

From 55% to 82% compliance. Whole team, sleeves rolled up, obsessed with the details, every function pulling on the same rope. 🤖❤️📈

At Norbert Health, we started with a simple mission: bring recurring monitoring and preventive care to post-acute facilities that desperately need it — but can't hire the staff to run it. It's been working. For residents. For their remote physicians. And now, we're turning our attention to the people at the heart of these facilities: nurses. Nurses carry a heavy load. The world is lacking about 40% of nursing resources, meaning that two nurses do the work that was designed for 3 nurses... Much of their day is consumed by recurring, repetitive tasks — necessary, but not always requiring human skill. Tasks that steal time away from what nurses actually want to do: talk to their patients. Care for them. Be present with them. One example that illustrates this perfectly: the Q15 protocol. When a patient experiences a mental health event or a fall, a strict protocol kicks in. Every 15 minutes, a nurse must check on that patient — logging their position, location, activity, and alertness level — for the next 24 to 72 hours. On paper, or in an app. Every. 15. Minutes. On a typical 50-bed nursing floor, there's almost always at least one patient on Q15 at any given time. And that single patient consumes a third of a nurse's shift — including nights. This is exactly what Norbert was built for. Precise, reliable, tireless. Our robots can handle Q15 monitoring seamlessly — visiting patients on exact intervals, logging observations automatically, and alerting nurses only when something is actually wrong. 24/7. Without burdening anyone. So nurses can do what only humans can do. 💙 Spend real, quality time with their patients. We built Norbert to take care of people. That includes the people who take care of people.

🤖 The robots winning in factories might not win in hospitals. Two bets. One really big.  But only one is right. 💰 Some companies have raised hundreds of millions of dollars to build general-purpose humanoid robots, perfecting them through sim-to-real training, making them as generic as possible, then shipping them to the world to do any job a human can do. 💡 Ambitious? Absolutely. A big bet - for sure. This is called the "Generalize first, then deploy" approach, as Matthieu Masselin told me recently. But in healthcare, this is the wrong bet. 🏥 Creating the robots that will help alleviate the shortage of over 1 million nursing professionals in the US — and make our healthcare system sustainable — will follow a different path: deploy first, then generalize. Here's why. Healthcare floors are uniquely complex. Patients, nurses, doctors, assistants, carts, devices — everything happening in parallel, all the time. And unlike factories, humans will always be there. Robots in healthcare aren't replacing people. They're working alongside them. 🤝 That changes everything about how you build. In this environment, a robot's most valuable capabilities aren't about moving things or performing procedures. They're about observing, sensing, measuring, and interacting — reading body language, detecting changes in patient status, communicating naturally. These skills need to be decoupled from motion and mobility entirely, because they'll need to translate across many different robot form factors over time. At Norbert, we're building this way from day one. 🚀 🎯 Start with a well-defined, high-value use case. Get into real clinical environments fast. Learn how nurses work, what they watch for, what they wish someone else could handle. Then add capabilities — new skills, new tasks. The more skills you add, the easier it will be to add new ones. Until the robot can take on a large share of the repetitive, tedious work that burns out healthcare professionals. One thing to avoid is getting rabbit-holed in a single task robot profile. For that, keeping our eyes on the long term vision, at all times, while delivering like crazy on the short term tasks, is the way to go. The general-purpose healthcare helper-bot is the destination. But you get there by being excellent at one thing first, earning trust, and building from there. 💪 #HealthcareRobotics #FutureOfCare #Robotics #HealthcareInnovation #DFTG Jean N. Mickael Munier Jeff Cardenas Noam Ohana Alexis Houssou David Reger Alex LeBrun Yaron Joels Yossi Wolf