Y Combinator’s latest Demo Day unveiled a cohort starkly different from its predecessors, signaling a profound shift in the startup landscape. This batch leaned heavily into “deep tech,” presenting innovations that felt more akin to science fiction than immediate commercial products. Investors, while buzzing with excitement over the groundbreaking ideas, also noted a refreshing return to more grounded valuations, a departure from the frothy investment climate of recent years. This curated list highlights the ventures that captivated early-stage VCs, marking them as the buzziest and most promising deals of the batch.
Key Takeaways:
- **Deep Tech Dominance:** The latest YC cohort showcases a significant pivot towards foundational “deep tech” innovations, spanning advanced robotics, novel computing architectures, and sustainable infrastructure.
- **Grounded Valuations:** Investors observed a notable return to more rational and less inflated valuations compared to previous cycles, suggesting a maturing and more discerning investment environment.
- **Tackling Grand Challenges:** From addressing global energy shortages with nuclear-powered data centers to democratizing robotics and revolutionizing defense manufacturing, these startups are confronting some of humanity’s most complex problems.
Another Y Combinator Demo Day concluded recently, leaving the tech world buzzing with a distinct impression: this batch was unlike any other. While every cohort brings a fresh wave of entrepreneurial spirit, the startups presenting this season skewed emphatically towards what can only be described as “deep tech.” The air was thick with ideas that one investor aptly described as “like science fiction,” pushing the boundaries of what’s currently possible in computing, robotics, and sustainable infrastructure.
As is customary, TechCrunch reached out to a network of early-stage VCs, seeking their insights on the hottest startups from this batch – both their personal top picks and the deals generating the most industry chatter. Our resulting list compiles those ventures that garnered enthusiastic mentions from at least two investors, indicating a broad consensus on their potential and innovative edge. Beyond the sheer audacity of the technological concepts, a recurring theme emerged from investor feedback: valuations were significantly more grounded than in recent cohorts, suggesting a recalibration in the market that prioritizes tangible innovation and realistic growth projections over speculative hype.
These are the companies that stood out, alphabetically, promising to reshape industries and potentially redefine our future:
The Infrastructure Innovators: Powering Tomorrow’s World
The demand for compute power continues its relentless ascent, fueled by the insatiable appetite of AI. This section highlights startups tackling the fundamental challenges of energy, data transfer, and processing at their core.
Automarine
What it’s building:Nuclear-powered data centers floating at sea.
Why it’s a fav:The global energy crisis, coupled with growing local opposition to new data center construction, presents a critical bottleneck for technological progress. Automarine, co-founded by an MIT computer science and naval engineer and an MIT PhD in nuclear engineering, proposes an audacious solution: highly efficient data centers housed on barges at sea. This innovative approach leverages seawater for near-free, abundant cooling, significantly reducing operational costs and environmental impact. The startup’s ambitious roadmap includes a gas-powered pilot by 2028, transitioning to floating nuclear power ships by 2032. Their vision has already resonated deeply with the market, securing over $4 billion in customer interest through letters of intent, a testament to the urgent need for their solution and their perceived revenue potential. This substantial early traction has positioned Automarine as one of the highest-valued startups in this YC batch, according to multiple VCs.
Dipole Labs
What it’s building:Effective and energy-efficient high-speed optical networking hardware for AI data centers.
Why it’s a fav:In the high-stakes world of AI, every nanosecond counts. GPU clusters, the workhorses of modern AI, often waste valuable compute time waiting for data to traverse between chips. The culprit? The energy-intensive process within the networking layer where data is constantly converted from light to electricity and back again, generating significant heat and consuming vast amounts of power. Dipole Labs has engineered a groundbreaking optical switch that bypasses this inefficient conversion. By keeping data as light, it can travel directly and efficiently where needed, drastically reducing latency and power consumption. This innovation directly addresses a critical pain point for AI data centers, maximizing the return on investment for incredibly expensive GPU hardware and ensuring that no compute cycle is left idle.
Lamb Labs
What it’s building:Custom inference chips with hardcoded AI model weights.
Why it’s a fav:The inference phase of AI — where a trained model makes predictions — is a massive energy sink. Traditional AI chips expend immense power fetching model weights from memory, creating a bottleneck that limits efficiency. Lamb Labs, founded by an Imperial College London AI Ph.D. and an Oxford theoretical physicist, is pioneering a radical solution: “Model Processing Units” (MPUs). These ultra-efficient custom chips hardcode AI model weights directly into silicon, eliminating the memory-bandwidth bottlenecks that plague conventional architectures. This fundamental rethinking of chip design promises to dramatically reduce the energy footprint of AI inference, a crucial step towards sustainable and scalable AI deployment, particularly as AI models grow ever larger and more complex.
Parasma
What it’s building:Training human brain cells to one day power compute.
Why it’s a fav:In the relentless pursuit of more energy-efficient computing, Parasma dares to look beyond silicon. This audacious venture is exploring the potential of using human brain cells as a biological alternative to today’s power-hungry AI computing hardware. While still in its nascent stages and operating at the very edge of scientific possibility, Parasma represents the ultimate deep tech bet on future compute. Its premise challenges conventional paradigms, suggesting that biological systems could offer unparalleled energy efficiency for highly complex computational tasks, potentially unlocking a new era of ultra-low-power, advanced AI. Investors are drawn to the sheer scale of the vision and the potential for a transformative breakthrough, even if the timeline remains distant and highly speculative.
Robotics & Automation’s Next Wave: Bridging the Physical and Digital
From autonomous physical labor to intuitive robot control, these startups are pushing the boundaries of how machines interact with our world, promising to reshape industries and daily life.
Praxis AI
What it’s building:Collecting real-world data on which to train robots.
Why it’s a fav:One of the biggest hurdles in developing truly capable robots is the scarcity of diverse, high-quality real-world training data. Praxis AI addresses this critical gap by partnering with businesses to capture extensive video and behavioral data of humans performing various tasks. This raw footage is then meticulously transformed into structured, actionable training material for companies building and deploying robots. Already collaborating with publicly traded companies and having collected data in over 150 different environments, Praxis AI is positioning itself as an indispensable enabler for the burgeoning robotics industry. As businesses increasingly experiment with the optimal division of labor between humans and increasingly sophisticated AI-driven robots, the detailed insights provided by Praxis AI will become paramount.
Nori
What it’s building:Creating affordable robots to handle everyday tasks.
Why it’s a fav:The dream of an affordable, functional humanoid robot for the home has long been a holy grail in robotics. Nori, launched just six weeks ago, has already generated nearly half a million dollars in sales, a staggering figure that underscores the market’s hunger for accessible automation. Priced at approximately $1,600 – a fraction of competitors like Neo, which can cost upwards of $20,000 – Nori aims to tackle mundane household chores like cleaning and folding clothes. Users can intuitively operate their robot via a laptop app, bringing advanced robotics into everyday life. Nori represents a bold attempt to answer one of robotics’ most pressing questions: can we build an economical, capable at-home robot that genuinely assists with daily tasks, moving beyond mere novelty to true utility?
Cosmic Robotics
What it’s building:Autonomous robots that can lift heavy objects.
Why it’s a fav:The founders of Cosmic Robotics harbor an ambitious long-term vision: to build a city on Mars. Their immediate strategy, however, is grounded in developing robust robotic technology capable of performing heavy-duty construction tasks here on Earth. The company’s technology is already actively involved in installing solar panels across the U.S. and has secured an impressive $25 million in contracts extending through 2027. This dual focus highlights a practical pathway to their grander ambition: perfecting autonomous heavy-lift capabilities on Earth as a proving ground for extraterrestrial construction. With an eye on SpaceX’s timeline for Mars colonization, Cosmic Robotics hopes to begin an exploratory mission by 2028, aligning their terrestrial innovations with the next frontier of human expansion.
Waddle Labs
What it’s building:An API layer that writes robot control code.
Why it’s a fav:The tech world eagerly anticipates the “ChatGPT moment” for robots – a paradigm shift where complex physical tasks can be commanded with natural language. Waddle Labs is directly addressing this excitement, but with a unique approach. Instead of merely training foundation models on raw video or human teleoperation data, this startup utilizes a sophisticated layer of Large Language Model (LLM) agents to directly write and control robot code. Positioned as “Claude Code for robotics,” Waddle Labs, founded by Harvard graduates, promises to democratize robot programming. Their claim is compelling: by plugging any hardware into Waddle’s API, developers can describe desired robot actions in natural language, and its AI agents will autonomously generate executable control code, verify its success, and set up the robot in approximately 20 minutes. This could unlock unprecedented flexibility and accessibility for robotic applications.
Defense & Geopolitical Tech: Redefining Security
Innovation in defense is often driven by geopolitical shifts and the urgent need for agile, cost-effective solutions. This category features a startup responding directly to these evolving demands.
Isengard Industries
What it’s building:Locally producible, jet-powered strike and counter-drones.
Why it’s a fav:In an era of increasingly complex global security challenges, the ability to rapidly and cost-effectively deploy defense technology is paramount. Isengard Industries aims to disrupt traditional defense manufacturing by mass-producing jet-powered attack and counter-drones directly within allied countries. This strategy drastically reduces the cost compared to prime contractors building them in the U.S. Co-founded by a former Australian Army officer and a defense entrepreneur who previously scaled another Ukraine-focused drone startup to $60 million in revenue, Isengard has already achieved an impressive $10 million in revenue itself. The startup’s timely and strategic offering has generated significant VC buzz, contributing to one of the loftiest valuations in this YC batch, according to two different investors. Their model addresses both affordability and supply chain resilience, critical factors in modern defense.
Investor Sentiment and Market Dynamics: A New Era?
The consensus among early-stage VCs underscores a fascinating dichotomy: the technology presented truly felt like a leap into the future, pushing scientific and engineering boundaries, yet the financial expectations were notably more tempered. This “science fiction” quality, married with “grounded valuations,” suggests a market that is both hungry for transformative innovation and wary of overinflated promises. It signifies a potential shift towards investing in ventures with strong technical foundations and clear, albeit often long-term, pathways to commercialization, rather than relying solely on market hype or rapid user acquisition metrics. This YC batch might just be a bellwether for a new era of tech investment, one that celebrates audacious engineering and foundational science while demanding financial prudence.
Bottom Line
Y Combinator’s latest cohort unequivocally demonstrates that the frontier of technological innovation is expanding rapidly, driven by bold entrepreneurs tackling the most fundamental and pressing challenges facing humanity. From redefining the very essence of compute with biological systems to building sustainable infrastructure on Earth and beyond, these startups are not just creating new products; they are laying the groundwork for entirely new industries. This batch, characterized by its deep tech focus and a refreshing return to grounded valuations, signals a maturation in the startup ecosystem. It’s a clear indication that the future of tech will be built not just on software and fleeting trends, but on enduring scientific breakthroughs and engineering marvels that promise long-term, transformative impact.
When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.
{content}
Source:{feed_title}