Mechatronics Technician

Can single-video learning solve the biggest obstacle in commercial robotics? Les-Leigh A August 27, 2026 What if a robot could learn to do a job just by watching a ten-second video clip? Skild AI reckons its new S1 foundation model can pull off exactly that, tackling unfamiliar physical tasks without needing a single round of fine-tuning. Built on a large manipulation dataset, S1 adapts on the fly with no retraining, no tedious data collection and no post-deployment calibration required. If these claims survive the real world, it would conquer one of the biggest challenges in commercial robotics. Reprogramming hardware for every new task is slow, eye-wateringly expensive and requires a room full of specialists. A machine that learns simply by watching changes the economics math of it all. What S1 can do (and what still needs proving). At its heart, Skild is showing off one-shot imitation learning straight from video. S1 studies a brief clip of a human or robot doing a task, then builds a custom execution policy without needing any extra training. By pairing visual inputs with natural language context, its video-language architecture figures out what is happening on screen and turns it into real-world robotic movement. Skild released benchmarks showing S1 beating current state-of-the-art models on generalisation tasks, even when faced with workflows completely excluded from its training. The company's demonstrations include manipulation tasks across different object types, lighting conditions and surface textures. CEO Deepak Pathak, who previously led robotics research at Carnegie Mellon and Meta AI, has framed S1 as a step toward robots that can be launched in any environment a human can describe or demonstrate. These results come from Skild's internal testing, which is standard for an early launch, but it does mean independent labs haven't verified the claims just yet. The robotics research community has seen confident one-shot learning announcements before that performed well in curated demonstrations, but less so in the full range of real-world conditions. Transitioning from controlled manipulation benchmarks to active implementations in warehouses, hospital wards or construction sites involves real-world friction. Benchmarks intentionally strip away the unpredictability these real environments require would. Why one-shot learning is A big deal. Traditionally, teaching a robot means gathering hyper-specific data, running heavy supervised or reinforcement learning and locking in a model that only works in one set of conditions. Want to switch tasks? You are starting back at square one. That's why factory robots stick to one assembly line, logistics hardware stays on fixed paths and dynamic automation is wildly expensive for most operators. Large foundation models have shifted the trajectory for language and image processing. By training one large model on varied datasets, it learns broad representations that easily adapt to tasks it wasn't explicitly taught. Skild is now bringing this play to physical robotics: feed S1 enough varied interaction data so that it develops transferable skills, letting it master new workflows through simple demonstrations instead of costly retraining. Dataset scale is everything here. Skild's claim of having the largest manipulation dataset is important, because foundation model generalisation lives and dies by data diversity. Train on a narrow set of tasks, the robot stays narrow. Expose it to a wide spread of hardware types, objects, environments and movement structures - and it actually has something to draw on when facing something new. The real question is whether Skild's dataset is truly rich enough to deliver the seamless adaptability they have promised. The market impact. Reliable one-shot video learning radically lowers both the cost and complexity of hardware implementation. An operator who can teach a machine a new workflow just by filming a quick demonstration opens the door to environments that couldn't previously justify the high setup costs. Small-batch manufacturing, flexible warehouse setups, commercial kitchens and physical retail are all spaces where traditional robotics economics fell flat. The robotics startup space has been building toward this. Physical Intelligence, acquired by Google in late 2024, was working on similar generalisation problems. Figure AI, 1X Technologies and Agility Robotics are all investing heavily in manipulation capabilities. Skild's approach of publishing a foundation model instead of building a specific hardware product positions it differently - as infrastructure for the robotics industry, not just a robotics company itself. For now, S1 is an impressive research showcase. The road from a solid benchmark to a rock-solid industrial release is notoriously long, and the robotics sector is infamous for overpromising on timelines. What makes S1 worth tracking is whether scaling foundation models is the right blueprint for how the problem ultimately gets solved.

Skild AI has raised $1.4 billion in its Series C funding round, bringing its total funding to over $2.2 billion and tripling its valuation to more than $14 billion in seven months. This marks the largest robotics AI funding round in history. The Pittsburgh-based startup, founded in 2023 by Deepak Pathak and Abhinav Gupta, emerged from stealth in July 2024 with a $300 million Series A at a $1.5 billion valuation. The latest round was led by SoftBank, with participation from NVIDIA's NVentures, Bezos Expeditions, Sequoia Capital, LG, Schneider Electric, and Salesforce Ventures. Skild AI develops a hardware-agnostic AI foundation model that can operate across various robotic platforms. The company acquired Zebra Technologies' Robotics Automation business in April 2026 and grew revenue from zero to approximately $30 million within months in 2025.

ABM and LaGuardia Gateway Partners launch autonomous robotics pilot at Terminal B. Robotic deployment includes autonomous floor scrubbers, autonomous vacuums, and one of the first autonomous robotic dogs in a U.S. airport terminal. ABM announced today the launch of a robotics program at LaGuardia Airport's award-winning Terminal B. In partnership with LaGuardia Gateway Partners (LGP), the operator of Terminal B, ABM is introducing both autonomous inspection and cleaning robots - including one of the first robotic quadruped "dogs" to be deployed in a U.S. airport terminal. The pilot program is the latest step in ABM's long-standing collaboration with LGP to deliver world-class guest experiences and operational excellence. Terminal B has already achieved global recognition as the first terminal in North America to earn the prestigious 5-Star Rating from Skytrax and was named the "World's Best New Airport Terminal" in 2023. Building on the successful deployment of the ABM Performance Solutions integrated facilities model and ABM Connect(TM) for Aviation, which leverage sensor data, IoT, and AI to optimize operations, the ABM robotics deployment will further enhance safety, efficiency, and the passenger experience across Terminal B. Robotics in action at Terminal B. The pilot features three advanced robotic platforms, designed to complement and support ABM's human workforce and elevate the terminal environment: * Robotic Dog - In partnership with Skild.ai, ABM is deploying a four-legged inspection robot, marking one of the first appearances of a robotic dog in an American airport. Skild AI is building a general-purpose robot brain for any robot morphology and task. Passengers may see the robot inspecting airport facilities, where it will quietly and efficiently support ABM staff in maintaining a safe, clean, and welcoming space. * Autonomous Floor Scrubbers - Complementing the inspection role of the robotic quadruped dog, these purpose-built scrubbers, in partnership with CenoBots, leverage advanced 3D LiDAR navigation and intelligent mapping to deliver consistent, high-quality floor cleaning. With the ability to run up to six hours autonomously, automatically recharge, and minimize downtime, the scrubbers help ABM redeploy staff to higher-value guest-facing tasks, while ensuring Terminal B continues to set the standard for cleanliness in the industry. * Autonomous Vacuums - Complementing the floor scrubbers, these dual-function autonomous units, deployed in partnership with CenoBots, use advanced navigation and intelligent mapping to capture both fine dust and larger debris across high-traffic terminal areas. Designed for continuous operation with self-charging capability, they help reduce manual effort, improve cleaning consistency, and enable ABM team members to focus on higher-value, guest-facing tasks, while maintaining Terminal B's industry-leading standard of cleanliness. Together, these platforms demonstrate how robotics, AI, and ABM Connect for Aviation work in unison to enhance facility performance, deliver measurable ROI, and create meaningful improvements in the passenger journey. "Airports are among the most dynamic environments in the world, and Terminal B is the perfect stage to demonstrate how robotics, AI, and data integration can transform facility operations," said Sean Bromfield, President of Aviation, ABM. "This pilot underscores ABM's leadership in anticipating our clients' evolving needs and investing in real, ROI-driven innovation. Robotics and AI are not about replacing people but empowering them - freeing our teams to focus on the work that most directly impacts travelers and strengthens the guest experience." "LaGuardia Terminal B has become a global benchmark for excellence in guest experience, and this robotics pilot takes our innovation commitment to the next level," said Suzette Noble, Chief Executive Officer, LaGuardia Gateway Partners. "We are proud to partner with ABM to test emerging technologies that align with our vision for a smarter, safer, and more seamless airport journey."