Building Deep Tech with Ilir Aliu

In this episode, I talk with Harald Schäfer, CTO at comma.ai, where he is leading one of the most interesting autonomy efforts in the world:They work on end to end driving and generative world models is changing how small teams can compete with billion dollar labs.We talk about his path from electrical engineering in Belgium and Santa Barbara to joining comma as one of the earliest engineers. Harald explains how he helped turn a hacker project into a focused engineering team that ships reliable autonomy to thousands of real users.He walks me through comma’s move to a single neural network that controls the car from video input, why deleting code is often more powerful than adding more, and how his team uses world models to train on billions of synthetic miles that never existed on real roads. Harald also shares what it is like to build inside a company with no CEO, why simplicity beats complexity in autonomy systems, and how the new comma 4 and Body 2 signal a move beyond cars into general robotics.If you work in robotics, autonomy, or AI systems, this conversation is packed with lessons about engineering clarity, avoiding brittle stacks, and shipping real products with small teams.

Albane Dersy turned down Goldman Sachs to build Inbolt, a robotics company now deployed in factories across the world. Her story is a masterclass in execution:In this episode, we talk about how Albane grew up in Paris, pushed her way through the French prep school system, and found her path into entrepreneurship after a semester at Wharton opened her eyes to what was possible. She explains how she met her two co-founders during the X EC program, and how the first version of Inbolt had nothing to do with robots. They started with a real-time guidance tool for workers and later pivoted to industrial robots after spending months on factory floors and seeing where customers really needed help.Albane walks through what it takes to sell and deploy automation inside global companies. She talks about why founders need to be on site all the time, and why selling early matters more than waiting for perfect reliability. She explains why deployment is everything in manufacturing and how Inbolt built a system that retrofit existing robots, reduced downtime, and proved value in a few weeks instead of years.We also talk about ambition, hard work, and the pressure she faced breaking into industries that are not always welcoming to young founders. Albane shares her early years in boxing gyms, her drive to be taken seriously, and the mindset that helped her operate and grow a company that now works with some of the biggest manufacturers in the world.If you want a clear look at how real robotics gets deployed at scale, and what it takes to build a company inside the most demanding industry in the world, this is an episode you should hear.

Dr. Hendrik Susemihl, CEO and Co founder of GoodBytz, shows you how fully automated kitchens can solve the labor crisis in food service and still serve better, fresher food at scale.We talk about his path from taking apart PCs as a teenager, to building large automation systems at Fraunhofer, to becoming CTO at NEURA Robotics. Hendrik explains why he walked away from a safe leadership role after his father’s heart attacks, how going plant based changed how he sees food, and why he became obsessed with the question: if I can cook healthy meals quickly at home, why is it so hard to get that quality in hospitals, canteens, and on the road.Hendrik breaks down how GoodBytz works in practice: a compact robotic kitchen that cooks up to 150 meals per hour, runs 24/7, and delivers consistent quality in places like university hospitals and motorway sites. We get into what they learned from running their own Lieferando brand, why he mostly ignores CVs and hires for people who build things for fun, and how a small Hamburg startup ended up signing a landmark contract with the US Army to feed soldiers in South Korea.If you care about robotics with real deployment, food at scale, or building a deep tech company that actually ships, this episode will be very useful for you.

In this episode, I talk with Jon Miller Schwartz, co-founder and CEO of Ultra, about how to actually get robots deployed in warehouses:We walk through Jon’s journey from tearing apart electronics on a tiny New York City workbench to Harvey Mudd, early YC startups in 3D printing, and building one of the first highly automated factories at Voodoo Manufacturing. Jon explains why those painful years with “last generation” robots convinced him to start Ultra and focus on one thing first e commerce order packing as a beachhead for real industrial deployment.He breaks down how Ultra’s robots drop into existing pack stations, learn from examples instead of brittle scripts, and why he believes in multi purpose robots before truly general purpose systems. We talk about force sensitive dexterity, what most people get wrong about warehouse automation, and how a small team in Brooklyn already has robots running live for customers. If you care about turning AI and robotics into shipped systems instead of slideware, this one is for you.

Axel Peytavin, co-founder & CEO of Innate, shows you how to teach real robots with language and quick demos without being a roboticist.We talk about Axel’s path from France to Stanford and why he is betting on personal robotics you can program with prompts, code, and demonstrations. He explains Mars, Innate’s $2K teachable robot with a Jetson Orin Nano, RGB-D vision, wrist camera, 2D LiDAR, and a 6-DOF arm.We break down BASIC, their open embodied agent that plans, remembers spaces, and chains skills. You will hear how a new skill can be trained in under 30 minutes, runs locally, and can be shared across a fleet.Axel walks through real use cases like chess play with camera understanding, pick and place, tidying, and security patrols. We cover the SDK, the open platform approach on ROS2, and why Innate focuses on accessibility, teachability, and community. If you want an insight into his story, a clear playbook for getting hands-on with embodied AI, and moving from lab demos to working robots, this episode is for you.

Tom Zhang, founder and CEO of Daxo Robotics: with over 100 actuators they challenge everything we thought we knew about dexterity.In this episode, we talk about his journey from growing up in a mountain village in China to launching one of the most talked-about robotics startups of 2025.Tom shares how early life on a family orchard shaped his fascination with building and problem-solving, what he learned during his years at Cornell and the University of Pennsylvania’s GRASP Lab, and why he believes the robotics industry has been climbing the wrong mountain by chasing simplicity instead of embracing complexity.We explore the story behind Daxo’s “Muscle v0” hand, how it was built in days with 108 tiny motors and off-the-shelf materials, and why redundancy, not minimalism, might hold the key to human-level adaptability. Tom also talks about his earlier success in agricultural robotics, raising over a million dollars in pre-seed funding, and what it takes to pivot from apple orchards to general-purpose robot dexterity.If you’re interested in robotics, entrepreneurship, or the mindset of founders who challenge fundamental assumptions, you’ll want to hear this conversation with Tom.

Maximilian Schilling, co-founder and CEO of warmwind, is building a new kind of browser where AI works like a digital employee: clicking, typing, and navigating apps visually instead of through APIs. In this episode, we talk about his mission to make automation transparent, reliable, and accessible for every business, and how he’s building one of Europe’s most ambitious AI startups from Jena, Germany.We dive into Max’s story, from growing up in a family that both inspired and warned him against entrepreneurship, to starting his first business at 14, and selling his second before launching Warmwind. He shares how financial independence as a teenager shaped his drive, why failure never felt like a real risk, and how curiosity (not comfort) has guided every decision he’s made.Max also explains Warmwind’s approach to building Warmwind OS, a browser-based system where AI agents automate workflows for small and medium businesses by acting on the screen instead of behind closed APIs. We talk about building reliable software, hiring in Europe, retraining vision-first AI models, and why he believes European founders should channel their “rage to compete” into world-class products.If you’re interested in AI, automation, or the mindset behind building bold companies from scratch, you’ll love this conversation with Max.

Robots still need weeks of coding to learn one new task. Xavier (Tianhao) Chi is changing that with Mbodi AI:Mbodi helps industrial robots learn through language and demonstration. No coding, no engineers, just simple instruction. We talk about how his team is closing the gap between advanced AI research and real factory floors, and what that means for the future of automation.Xavier shares his path from growing up in Shenyang to leading Google Public DNS, one of the internet’s core services, and why he left to build Mbodi with his co-founder. He explains why the next wave of robotics will come from adaptable software, not humanoids.We also talk about risk, ambition, and what it takes to move from stable engineering to startup chaos. Xavier breaks down Mbodi’s hybrid AI approach, its sub-0.5 second response times, and how their partnership with ABB is turning it into real deployments.A must-listen for anyone building in robotics, AI, or industrial automation.

Ashish Kapoor is building General Robotics to solve the biggest deployment problem in robotics: Getting real robots to work in the real world. In this episode, he shares how he’s doing it, and why most robotics stacks aren’t built to scale.We talk about growing up in India, studying at IIT and MIT, and how his mindset shifted from solving hard problems to finding the right ones. Ashish shares why he left research to start General Robotics, the limits of today’s robotics stacks, and how Grid aims to solve the deployment bottleneck, especially for enterprises drowning in PoCs and fragmented software.He also opens up about his background in aviation, building his own airplane, and how he's betting on cloud-first skills infrastructure while others chase edge. This one’s packed with insight from someone who’s worked across every layer of the robotics stack... and is now trying to make it all work in the real world.

In this episode, I talk with Brian Walker, founder and CEO of REVEL, the company building the simulation backbone for humanoid robotics. Brian’s journey started far from Silicon Valley: growing up in the Czech Republic and working on Hollywood sets like Avatar and The Mandalorian, where he helped pioneer real-time XR production.We talk about what pulled him from filmmaking into robotics, and how sci-fi inspired him to stop watching the future and start building it. Brian shares why he founded REVEL to create a massive library of digital twins, turning real-world products into high-fidelity simulation assets so robots can train on them before ever touching them.His goal? To make every product “robot-ready” and compress a decade of physical experience into just hours of training.We also dive into his views on self-education, outsider thinking, and why he acquired a startup during a 20-hour hackathon, with a mic-drop and a €20K offer.If you’re into robotics, simulation, or stories of wild career pivots, don’t miss this one.

A Stanford physicist leaves academia to build open-source software for humanoid robots? I talked to OpenMind founder Jan Liphardt: OpenMind a new robotics company building an open-source, AI-native operating system for humanoid robots.We talk about being born in Germany, his upbringing in Michigan, early love for taking things apart, and how his path led from biochemistry at Reed to a PhD at Cambridge, then faculty roles at Berkeley and Stanford.Jan shares why he made the leap from academia to entrepreneurship, how a Nature paper and a Christmas Eve email nudged him out of the lab, and what drives his belief in transparency, modularity, and decentralized control for intelligent machines.We also discuss OpenMind’s strategy, where robots download their rulebooks from Ethereum, and why he thinks humanoids won't fold your laundry... but could teach your kids or assist in hospitals.

🎙️ I talked with Bob van Luijt, co-founder and CEO of Weaviate, the open-source vector database that's become core infrastructure for AI-native applications.We talk about how Bob grew up in a small Dutch town, started coding in QBasic, and built his first software company while still in school. Then came the unexpected turn: jazz. He shares how studying music (from a conservatory in the Netherlands to Berklee in Boston) taught him grit, deep focus, and how to think in systems. For Bob, writing code and playing music happen in the same part of the brain.We talk about how Weaviate began as a side project fueled by curiosity about the distance between words, and how that simple idea turned into one of the most used vector databases in the world. Bob explains how the release of transformer models unlocked everything, and how he's stayed focused on helping real developers build, not just chasing hype.We also get into his philosophy on building companies, how he thinks about talent and education, and why he believes too much "academic thinking" blocks real potential. Bob’s not in it for the ego or the exit... he’s building tools for other builders.

In this episode, I talk with Vikash Kumar, Adjunct Professor at Carnegie Mellon and founder of MyoLab.AI, where he’s building human-embodied AI systems:We talk about growing up in a small Indian town, the influence of his mother on his early learning, and how a robotics club at IIT Kharagpur set him on a 15-year path through the world’s top labs.From a PhD at the University of Washington, to OpenAI, Google Brain, Meta FAIR, and now his own company! Vikash shares how his curiosity evolved from tinkering with machines to uncovering the fundamentals of embodied intelligence, and why he believes the future of AI is physical, not just linguistic.He also explains the bold vision behind MyoLab: building “digital twins” that are physiologically and behaviorally lifelike; AI companions that understand not just what you say, but who you are. We talk about how this intersects with robotics, health, memory, and agency, and why the path to general intelligence may start in the body, not the cloud.

I talked with Brennand Pierce, founder and CEO ofKinisi Robotics, where he’s building one-armed mobile manipulators:Designed to automate warehouse tasks like picking, palletizing, and labeling and many more.After nearly two decades in robotics, Bren brings a rare mix of academic depth, startup experience, and hands-on engineering to the conversation.We talk about growing up fascinated by sci-fi and Japanese hobby robots, studying computer science at Exeter, earning a Master’s at the Bristol Robotics Lab, and completing his PhD in humanoid robotics at the Technical University of Munich. Bren shares what he learned building humanoids, founding three robotics companies; including co-founding Bear Robotics, which shipped over 25,000 service robots, and why he now believes the future belongs to practical, task-optimized robots rather than overpromised humanoids.We also look into Kinisi’s approach to solving real-world deployment challenges, lessons from past robotics booms, and what it takes to move from flashy demos to robots that actually work in production.

In this episode, I talk with Aaron Tan, PhD, Co-founder of Syncere, where he’s reimagining domestic robotics:A robotic lamp capable of folding laundry!One single video of the concept went viral, pulling in over 4 million views and sparking thousands of conversations online.We talk about growing up in Taiwan, immigrating to Canada as a teenager, and how a Lego Mindstorms kit from his father kicked off a lifelong obsession with robots. Aaron walks me through his path from a Bachelor’s and Master’s at Ontario Tech University, to a PhD at the University of Toronto, and a postdoc at Stanford... and why he chose the Bay Area to build Syncere.He also shares the story behind Lume’s design, inspired by Beauty and the Beast, and explains why he believes the future of home robotics shouldn’t look like humanoids. We talk about building in public, customer psychology, and what it’s like creating a product people can’t wait to have in their homes.

In this episode, I talk with Brendah Njiru, founder of HOMY Robotics, where she’s building emotionally intelligent humanoid robots for senior living:With a background in neuroscience and Alzheimer’s research at Cornell, Brendah brings a rare scientific depth to robotics.We talk about her upbringing in Kenya, her early obsession with medicine, and what pulled her into AI and hardware. She shares how she transitioned from labs to startups, why senior care is the perfect proving ground for home robotics, and how her work is grounded in real-world deployment.Brendah also opens up about pressure, ambition, and how she's building HOMY to solve deep human problems, not just automate tasks.

In this episode, I talk with Lerrel Pinto, Assistant Professor at NYU and one of the most cited researchers in robotics today:His work spans everything from self-supervised learning to robot dexterity, and he's on a mission to make robots generalize the way humans do.We talk about growing up in India, building his own education at IIT, and what led him to Carnegie Mellon, Berkeley, and now New York. Lerrel explains how his lab at NYU, GRAIL, tackles robot learning at scale (from representation learning to reinforcement learning) and why open-source, affordable robots are core to his approach.He also shares what it’s like launching his new stealth-mode startup, Assured Robot Intelligence (ARI), while running one of the top research labs in the country. We talk about how his teaching, mentoring, and outreach are shaping the next generation of roboticists.

I sit down with Benji Barash, CEO of Roboto AI, to talk Amazon, robotics, and why knowing what to do with your data is key to scaling:Benji spent years at Amazon working on drone delivery, but started to see a growing problem in robotics, the data was piling up faster than teams could make sense of it.He left Amazon to build a solution. Today, Roboto AI helps robotics companies analyze massive amounts of sensor logs and time-series data. It’s like a copilot for engineers trying to figure out why something broke, how to improve it, and what to do next.We talk about growing up in the UK, getting into programming way before school even taught it, and what it takes to go from big tech to a lean startup. Benji shares what surprised him about building in the real world, how he works across time zones with his co-founder in Zurich, and what it means to build tools that help others scale.

In this episode, I talk with Madison Maxey, founder of LOOMIA, a company building soft, flexible electronics for everything: Everything? Everything! From robotics to automotive interiors. Maddie’s journey spans fashion school, a Thiel Fellowship, a return to Stanford in her mid-20s to study material science, and a decade of turning prototypes into real-world tech.We talk about growing up with a soldering iron and a sewing machine, how she designed a smart jacket for Zac Posen and Google, and why building a company means more than building a product. Maddie shares what it took to land early customers like Airbus, how she balances long timelines with fast-moving industries, and why her goal is to build something meaningful over 30 years, not just raise another round.We talk about early wins, hard lessons, the beauty of tactile sensing, and why confidence comes from doing hard things until they start to feel normal.

Yesterday, they launched a $299 robot. It looks like a toy, but it opens up a world of AI. Today, I talk to the person who made it real.Matthieu Lapeyre is the founder of Pollen Robotics and one of the most quietly influential roboticists in Europe. He’s been building open-source humanoids long before it was cool, from Poppy to Reachy to Reachy Mini.We talk about how growing up without a tech background shaped him, why he left research to ship hardware, and how he kept going through years of bootstrapping with barely enough to pay the team. He shares what it’s really like to live on the edge for years, and how joining Hugging Face gave them the launchpad they needed.We also get into the making of Reachy Mini, why it's designed to be unbreakable, what inspired the egg-shaped head, and how it could become the iPhone moment for robotics.