From Moon Hotels to cattle herding: Investors chased 8 startups at YC Demo Day

Investors have flocked to Y Combinator’s Demo Days for years to get their hands on promising startups building great technology. After all, this accelerator has produced some of the world’s biggest tech companies, from Airbnb and Reddit to Dropbox, Zapier, and Stripe.

That’s why we make sure to monitor the event to discover the most interesting companies from each batch. As I have done almost every quarter since the accelerator moved to four cohorts per year, I asked about a dozen investors which startups were in the most demand at Y Combinator’s Winter 2026 Demo Day earlier this week.

To ensure that our list includes truly desirable featured companies, a company had to be marked as a “preferred company” by at least two different venture capital investors to make the cut.

As for valuations, I hear that at least two startups have raised $100M in funds, although these startups are already generating revenue with a run rate of $1M or more. Even for lesser-known startups not on this list, the “hypothetical” valuation for this quarter appears to be around $30 million, which investors tell me is roughly double the current seed market average.

Without further ado, here’s the list:

Beyond Reach Laboratories

What it builds: Deployable solar arrays for satellites.

Why is it preferred: The startup claims to have developed solar arrays that are the size of a dining table at launch, but unfold to the size of a football field when they reach orbit. The founders say their system can increase available energy tenfold while cutting costs by 88%. Beyond Reach already has a flight planned for 2027, and says it has secured letters of intent worth $325 million from leading space companies.

byteport

What it builds: FTP is ridiculously fast.

Why is it preferred: According to Byteport’s founder, Jayram Palamadai, current file transfer protocols like TCP are too slow for the age of AI. That’s why he built DART, short for Dynamic Accelerated Record Transfer, which can apparently transfer large files 10 times faster than TCP, and even up to 1,500 times faster on “reliable connections.”

Hexagon security

What it builds: AI-powered continuous security testing tools.

Why is it preferred: To combat hackers who use AI to launch persistent cyberattacks, Hex is building AI agents that can act as penetration testers, constantly scanning for vulnerabilities and vulnerabilities in companies’ infrastructure. By automating what was previously an infrequently performed manual process, Hex claims it can prevent attacks at a fraction of the cost. The startup claims to have surpassed run-rate revenues of more than $1 million in just eight weeks, which may be why VC investors are “struggling,” as one person told me, to invest in the company.

Grazymes

What it builds: Autonomous drones for livestock grazing and monitoring.

Why is it preferred: Transporting livestock on large farms is an expensive and dangerous task, often requiring helicopters and motorcycles. The GrazeMate founder, who grew up on a 6,000-head cattle station in Australia, saw a way to make life easier for ranchers, so he dropped out of college where he was pursuing a degree in robotics.

GrazeMate drones can automatically guide livestock to different areas of the farm, estimate the animals’ weight, grass availability and growth, and can follow pre-determined route plans.

Puppy space

What it builds: Permanent lunar infrastructure, starting with a hotel on the moon.

Why is it preferred: “Humanity will go interplanetary,” says Skyler Chan, founder of GRU Space, a recent Berkeley graduate who previously built software at Tesla and worked in NASA-funded space technology. “It’s a question not of if, but when, and the time is now.”

Chan claims his startup has developed a “moon factory” that can turn lunar soil into structural bricks, which he plans to use to build a luxury hotel on the moon as a “wedge” for the wider lunar infrastructure. GRU’s astronomical aspirations, including its goal of opening the first hotel on the moon by 2032, have made it one of the most talked-about startups in this YC group. The company has already secured $500 million in letters of intent, an invitation to the White House, and even a booking from the Trump family.

Lowell

What it builds: A marketplace for human-captured data for multimodal AI training.

Why is it preferred: The Lowell company, founded by two UC Berkeley dropouts, is building a data marketplace that connects AI model makers with contributors who can submit “everyday life” activities, such as ironing or conversations between a patient and a doctor, to provide audio, video and image data. The company claims to have achieved an ARR of nearly $2 million within six weeks, fueled by high demand from robotics and voice AI labs.

Pax Historia

What it builds: An alternate history strategy game powered by artificial intelligence.

Why is it preferred: Pax Historia allows users to rewrite history in a way that traditional strategy games cannot. Using generative AI, the game responds to countless complex geopolitical scenarios, from “What if Rome had never fallen?” to “What if the United States took Greenland?” The founders claim that the game currently attracts 35,000 daily users who have played nearly 20 million rounds.

Stelta

What it builds: Artificial Intelligence Agent for Intellectual Property and Patent Lawyers.

Why is it preferred: Stilta’s founders claim that patent disputes can cost up to $4 million per case, largely due to the costs of manual document review. The startup says its AI agent can search and analyze patents across databases and scientific literature, saving time and legal fees.

The company’s agents are already being used by intellectual property lawyers at pharmaceutical giant Roche. For investors, another attractive aspect is that the founders hail from Sweden – recent Swedish successes such as Lovable and Legora have created something of a “halo effect” around companies from the region, one venture capital investor said.