Participated · Equity · Mar 2026

RoboForce develops scalable Physical AI-powered robots and a robot foundation model designed for demanding industrial work that is repetitive, hazardous, or safety-critical. The company integrates real-world fleet data and high-fidelity simulation to create a closed-loop data flywheel intended to compound intelligence across deployments. RoboForce partners with NVIDIA for edge compute and simulation tooling, citing NVIDIA Jetson Thor, Isaac Sim, Isaac Lab, Cosmos, and OSMO in its stack. It targets sectors such as utility-scale solar, data centers, mining, shipping, manufacturing, and logistics and is transitioning active pilots toward production deployments. Financially, RoboForce raised $52 million in the latest oversubscribed round, bringing total funding to $67 million. The company was founded in 2023 and is headquartered in Milpitas, Calif.

Participated · Series A · Jul 2024

Skild AI is developing the Skild Brain, a scalable foundation model that can control quadrupeds, humanoids, tabletop arms, mobile manipulators, and virtually any machine capable of movement without needing prior knowledge of the robot’s exact body form. The technology enables robots to tackle tasks that range from routine household chores—such as cleaning, loading a dishwasher, or cooking an egg—to demanding assignments like navigating slippery terrain. In 2025 the company leapt from zero to roughly $30 million in revenue within a few months, demonstrating early commercial traction as it rolled out solutions for security, facility inspection, last-mile delivery, and warehouse automation. The newly raised capital will be directed toward enlarging model-training capacity and accelerating enterprise and commercial deployments. Headquartered in Pittsburgh with offices in the San Francisco Bay Area and Bengaluru, India, the team is led by CEO and co-founder Deepak Pathak and President and co-founder Abhinav Gupta. Strategic partners such as Samsung, LG, Schneider Electric, and Salesforce Ventures signal strong corporate interest in embedding the Skild Brain across varied hardware platforms. With an omni-bodied approach and rapid revenue growth, Skild AI aims to become the default intelligence layer for the robotics industry.

Participated · Seed · Jan 2024

ViralMoment provides an AI platform that analyzes short-form social videos end-to-end to deliver insights, strategy, and analytics for brands and agencies. The platform goes beyond captions, transcripts, and thumbnails to offer a complete view of a brand’s reputation and viral trends in real time. Customers include WPP, Omnicom, and Warner Brothers. The company has tracked over 1 billion social videos to date. Led by CEO Chelsie Hall and based in Menlo Park, CA, ViralMoment plans to use new funding to widen the scope of its video-reading AI. The company recently raised $2.5M in Seed funding.

Participated · Seed · Oct 2023

Aug X Labs develops Augie, an AI-assisted video creator that converts prompts, scripts, narration, recordings, or clips into complete videos and supplements content with commercially licensed stock footage. Augie runs in any browser today and a dedicated mobile experience is planned. The platform’s content-matching system combines computer vision, metadata, GPT, and NLP and integrates technologies such as ElevenLabs (AI voices/voice cloning), Replicate.AI, AWS, OpenAI, and Hugging Face. The company plans to use the new funding to hire and retain staff, continue core technology development, and expand customer acquisition as Augie exits beta later this year. Operating traction reported in the article includes double-digit growth since open beta and power users creating an average of 37 Augies per month. No revenue, founding year, or location details were provided in the article. Aug X Labs is developing an AI-based visual storytelling platform that automatically converts text or audio prompts into dynamic, purpose-driven videos. The platform lets users augment video creations with open or licensed media, generative media (AI art), and owned media. It targets creators who currently face expensive, time-consuming tooling or the need to hire professionals, aiming to democratize video production so anyone can produce engaging visuals. Anyone with an iPhone or webcam can record and upload video to the system, which is designed to simplify creation and lower the skill barrier. The company plans to reveal its product later this year as it enters public beta. Aug X Labs announced a $1.0M pre-seed funding round to support development and product launch.

Participated · Seed · Aug 2023

Refiberd builds hyperspectral imaging and AI technology to identify textile fiber composition and enable accurate sorting for recycling. Its system can be integrated into existing recycling infrastructures so materials are processed with the correct methods and turned into new, usable fibers. The product targets the waste-sorting dilemma in fashion to advance textile-to-textile recycling. Refiberd was named eBay's Circular Fashion Innovator of the Year and received mentorship and industry access through the program. CEO Sarika Bajaj said the investment will help close a significant round, expand the technology to new recycling partners, and strengthen the company's AI capabilities. Refiberd is a US business. Refiberd builds hyperspectral imaging and AI systems to identify textile materials and blends to enable accurate sorting for textile-to-textile recycling. The company points hyperspectral cameras at garments, samples dozens to hundreds of bands of light, and uses neural networks plus generative AI to distinguish thousands of textiles and fill gaps in its dataset. Refiberd has assembled a sample library of over 10,000 entries and generates synthetic spectra to improve model accuracy for materials not present in its collection. One camera unit can eventually sort up to 7 million pounds of material, and the startup is targeting pilots with textile companies, chemical recyclers, mechanical recyclers, and textile sorters to refine its product. Founded three years ago by Sarika Bajaj and Tushita Gupta, Refiberd frames its work as a response to the roughly 14 million tons of clothing landfilled or incinerated annually in the U.S. The company will use its recent funding to roll out initial pilots and further develop its hardware and AI models.