Y Combinator’s latest Demo Day featured a group of startups with a strong emphasis on deep technology innovations. Early-stage venture capitalists highlighted nine companies that stood out for their ambitious approaches and potential market impact.
One notable startup, Atomarine, is developing nuclear-powered data centers on floating barges to address power shortages and community resistance to new land-based data centers. By leveraging seawater for cooling, Atomarine aims to launch a gas-powered pilot by 2028 and transition to nuclear-powered vessels by 2032. The company has reportedly secured over $4 billion in customer interest.
Dipole Labs is tackling inefficiencies in AI data centers by creating optical networking hardware that eliminates the energy-intensive conversion between light and electricity, improving data transfer speeds and reducing power consumption.
Isengard focuses on producing jet-powered attack and counter-drones domestically within allied countries at lower costs than traditional contractors. With $10 million in revenue and a leadership team experienced in defense startups, it has attracted significant investor attention.
Lamb Labs is developing custom inference chips called Model Processing Units that hardcode AI model weights into silicon, aiming to reduce energy use and overcome memory bandwidth limitations common in AI hardware.
Another company is collecting extensive real-world video data of human work activities to provide training material for robotics developers, supporting the growing interest in automating tasks best suited for robots.
Nori offers an affordable humanoid robot priced around $1,600, designed to assist with household chores like cleaning and folding clothes. This contrasts with more expensive alternatives and addresses the challenge of creating practical at-home robots.
A startup focused on autonomous heavy-lifting robots is already deploying technology to install solar panels and has secured $25 million in contracts through 2027. Their long-term vision includes automating construction for Mars colonization, aiming for exploratory missions by 2028.
Parasma is exploring the use of human brain cells as a novel, energy-efficient computing substrate to reduce the power demands of AI model processing.
Waddle Labs is developing an API that uses large language model agents to write robot control code, enabling developers to command robots in natural language. This approach could accelerate robotics programming by autonomously generating and verifying executable code.
This cohort’s focus on deep tech solutions addresses pressing challenges in energy, robotics, and AI infrastructure. The realistic valuations and strong investor interest suggest these startups could influence their respective industries significantly in the coming years.