AI was as soon as once more the buzzword for this newest batch of YC Demo Day firms. Practically 190 firms participated in Y Combinator’s Winter ‘26 cohort and introduced their startups in a Demo Day on Tuesday.
These firms are engaged on merchandise throughout industries corresponding to regulation, transportation, and healthcare.
I didn’t, admittedly, hear to each single product pitch given the sheer measurement of the cohort and this yr’s Demo Day format out there to media: YC posted the pitch movies, one after the other, round 20 minutes after their shows (moderately than a livestream, or an in-person session invite).
As an alternative, I examine all 190 of the startups presenting and spent the day watching pitches from these we discovered intriguing, then narrowed it right down to the 16 that stood out as probably the most fascinating startups of this overflowing YC class.
ARC Prize Basis
What it does: Creates benchmarks to assist measure progress towards AGI.
Why it’s fascinating: A nonprofit in YC! However then once more, when OpenAI, Anthropic, and GoogleMind are already utilizing some type of the group’s benchmarks, it is sensible why it will be included. This basis goals to encourage extra open-source AGI analysis by internet hosting competitions and awarding analysis grants. One motive for this AI revolution is to succeed in AGI (which Nvidia CEO Jensen Huang says has already arrived), and it is going to be a matter of historic file for monitoring how shut or we’re from AI machines that have a basic intelligence.
Asimov
What it does: Collects human motion information to coach humanoids
Techcrunch occasion
San Francisco, CA
|
October 13-15, 2026
Why it’s fascinating: Folks from all over the world submit movies of themselves performing actions and duties and to this firm so it will probably then flip it into datasets that may assist prepare robots. It’s a part of the motion attempting to make humanoids a factor, discovering makes use of for them past the availability chain and leisure. I’m bullish on humanoid expertise, despite the fact that our “Rosey the Robotic” period nonetheless could be eons away. Utilizing information to show humanoids the circulate and — dare I biasly say, magnificence — of human motion may assist them much less, properly, robotic as they carry out duties.
Avoice
What it does: Helps automate the tedious non-design work for structure companies
Why it’s fascinating: It’s not day by day you hear about new expertise focusing on the architectural business. The founders themselves famous that this market is underserved (although wealthy in potential). This device makes use of AI to assist automate duties that inventive varieties like architects my discover tedious like reviewing specs, drawings, contracts and proposals.
Button Laptop
What it does: A wearable AI
Why it’s fascinating: Everyone seems to be attempting to make wearable AI a factor because the world awaits OpenAI’s product from its acquisition of Johnny Ive’s firm. Two former Applers (can I name them that?) have joined collectively to launch Button, basically a tiny pc, the founders defined, that’s constructed for AI. Button connects to apps like e mail, Slack, and Salesforce and operates them through voice command to carry out sure duties. The following must-have {hardware} is more likely to be some type of AI wearable, so it’s fascinating to see what’s rising.
CodeWisp
What it does: Let’s anybody construct video games utilizing AI
Why it’s fascinating: The founders say all it’s important to do is inform an AI make a recreation, and it’ll make the sport. That’s enjoyable, inventive, thrilling! I attempted to construct video games as a toddler and at all times discovered it onerous and tedious, however the pleasure of imaging designing one by no means went away. Whereas vibe coding has turn into the trend for constructing apps, instruments like this make imaginative execution a lot simpler. Possibly this would be the subsequent era of vibe constructing.
Crosslayer Labs
What it does: Helps detect web site spoofs
Why it’s fascinating: The rise of agentic instruments implies that web sites have gotten simpler to spoof with the dangerous guys usually are not shying away from utilizing this tech to rip-off folks. Crosslayer Labs helps its prospects detect and monitor their on-line setups to allow them to keep protected towards this shade of rising web threats.
Doomersion
What it does: Teaches you languages as you doomstroll
Why it’s fascinating: We spend manner an excessive amount of time doomstrolling, simply filling our brains with slop that finally ends up both irritating us or atrophying our brains. This startup is an app that exhibits customers quick movies, like how they’d see and scroll by way of them on a TikTok feed, within the language they’re attempting to be taught. If there needs to be some that means to all of the content material we stuff into our heads all day, what a superb technique to discover that. It combines what shoppers gained’t cease doing (spending hours on their telephone swiping by way of content material) with one thing as cool as language studying. Très intéressant.
Lexius
What it does: Embeds superior AI into safety programs
Why it’s fascinating: This makes use of AI to improve current safety digicam programs, enabling footage to detect and report cases of theft or falls, changing a fragmented, primarily guide course of. The startup says its focused companies that have cameras with out AI intelligence, the place a digicam would possibly catch an incident, however the firm is delayed in taking any motion.
Librar Labs
What it does: An AI-powered library administration system device
Why it’s fascinating: That is AI touching a type of industries that the tech business usually overlooks — libraries. This startup has created an AI-powered library administration system to particularly assist, as of proper now, faculties with stock and cataloguing. Because the founder mentioned in his pitch, there isn’t a lot competitors on the subject of automating or innovating the instruments already used on this area, making any new concept a contender for “the subsequent large factor.”
Milliray
What it does: Radar system to assist monitor small drones
Why it’s fascinating: Protection tech is among the hottest classes in tech proper now. Because the founding father of this firm mentioned, proper now, persons are within the fields doing every part they’ll to trace tiny drones, however the human eye can miss a lot or mistake a tiny drone for a chook and vice versa. This startup makes use of sensors to establish what is definitely a tiny drone within the sky. Given the present state of geopolitics, new expertise is required — and rising — so that each nation can keep forward of the sport and its enemies.
MouseCat
What it does: Makes use of AI to analyze fraud
Why it’s fascinating: AI is a superb device for work, and that holds true for these whose work is fraud and scamming. This firm pulls an organization’s information from the big cloud storage like Databricks or Snowflake, analyzes shopper information and exercise for something suspicious, and offers suggestions on take motion. AI-native instruments like this are essential on the subject of maintaining with the dangerous AI that can also be able to unleashing.
Opalite Well being
What it does: Makes use of AI to assist healthcare suppliers discuss to non-English audio system
Why it’s fascinating: There may be a lot left as much as interpretation when two folks can not perceive one another. Within the medical world, it might be life or demise. This AI medical translator helps break the language barrier, enabling healthcare suppliers to know sufferers who converse a unique language. In a globalized world (and in a rustic as numerous as America), it’s essential for folks to have the ability to entry the healthcare they want, no matter language. In fact, this concept isn’t fully authentic, as plenty of different startups and healthtech suppliers provide an identical service.
Sequence Markets
What it does: Let’s folks commerce throughout varied markets, like crypto and prediction, on one system.
Why it’s fascinating: As somebody who likes every part in a single place, I perceive wanting a much less fragmented course of when executing trades in these markets. It’s the identical motive I nonetheless like going to big-box retailers to purchase manufacturers (I prefer to see all my choices directly).
ShoFo
What it does: A video library of principally every part
Why it’s fascinating: This startups is billing itself because the “world’s video library,” which is fairly cool should you ask me. I grew up on YouTube and Tumblr and keep in mind how delicate search was again then, if you had been searching for one thing particular. Although that is extra of a customized video index to assist AI Labs discover numerous datasets effectively, I really like any device that makes looking and organizing simpler.
Sonarly
What it does: Helps software program repair its personal manufacturing points
Why it’s fascinating: This startup is making tech that simply sounds very cool. It connects to different monitoring programs, guarantees to cut back alert noise (a distraction from discovering the alerts that truly matter), mechanically identifies the foundation causes of issues, after which finds methods to both repair them or counsel additional actions for the engineers. Whereas there are a rising variety of AI code assessment startups (and that characteristic can also be being provided by the mannequin makers), there needs to be room for impartial ones as soon as the code hits manufacturing programs. That is yet one more side of the workflow that founders are automating.
Terranox AI
What it does: Makes use of AI to seek out uranium deposits in North America
Why it’s fascinating: Uranium will likely be wanted to energy the subsequent era of nuclear power, the founders of this firm reminded us. (Nuclear poweris typically thought-about secure lately, although the uranium itself is, clearly, poisonous — and the founder didn’t embody within the pitch the way it might be safely excavated.) These founders consider that nuclear will likely be wanted to assist energy the entire new information facilities being constructed. Earth goes to wish loads of power, from many sources, to make sure its human inhabitants can sustain with the ambitions of the AI revolution.
