YC Winter 2026 Demo Day: 16 Startups to Watch 🚀
Y Combinator’s Winter 2026 cohort was another AI-forward batch: nearly 190 companies presented during Demo Day. I reviewed the pitches and materials and narrowed the list to 16 startups that stood out for their ideas, traction, or potential impact. Below are concise profiles of each company, what they do, and why they’re worth following.
ARC Prize Foundation
What it does: Creates benchmarks and competitions to measure progress toward artificial general intelligence (AGI).
Why it’s interesting: As a nonprofit inside YC, ARC Prize provides open benchmarks and grants that help academic and open-source AGI research — an important counterbalance as big players race toward general-purpose AI.
Asimov
What it does: Collects human movement video data to train humanoid robots.
Why it’s interesting: By crowdsourcing motion datasets, Asimov aims to teach humanoids the fluidity and nuance of human movement, expanding potential uses beyond warehouses and entertainment.
Avoice
What it does: Automates non-design, administrative work for architecture firms using AI.
Why it’s interesting: Architecture is an underserved vertical for productivity tools. Avoice automates tasks like spec reviews, contracts, and proposals so architects can focus on creativity.
Button Computer
What it does: A small wearable AI-focused computer that connects to apps like email, Slack, and Salesforce and performs tasks via voice commands.
Why it’s interesting: Wearable AI is emerging as a front-line interface trend. Button aims to be a compact device built specifically for hands-free, AI-driven workflows.
CodeWisp
What it does: Lets anyone build games by telling an AI what to create.
Why it’s interesting: Democratizing game creation could unlock a wave of creative indie titles and rapid prototyping, making game development accessible to non-programmers.
Crosslayer Labs
What it does: Detects website spoofs and monitors online assets to protect brands from fraud.
Why it’s interesting: As agentic systems and automated tooling grow, spoofed sites and phishing become easier to create. Crosslayer helps companies detect and respond to these threats faster.
Doomersion
What it does: A short-video language-learning app that teaches through the same swipeable feed format users already consume.
Why it’s interesting: It turns mindless scrolling into micro-lessons, combining addictive short-form content with language acquisition techniques.
Lexius
What it does: Embeds advanced AI into existing security camera systems to detect theft, falls, and other incidents.
Why it’s interesting: Many businesses have cameras but lack intelligent monitoring. Lexius turns passive footage into actionable alerts, reducing investigation time and improving safety.
Librar Labs
What it does: An AI-powered library management system focused on schools for inventory, cataloging, and search.
Why it’s interesting: Libraries and schools are often overlooked by tech innovation. Librar Labs brings automation and better discoverability to a fragmented, essential sector.
Milliray
What it does: A radar and sensor system to detect and track small drones.
Why it’s interesting: With small drones proliferating, human observers can miss or misidentify them. Milliray provides automated detection useful for defense, critical infrastructure, and events.
MouseCat
What it does: Uses AI to investigate fraud by analyzing customer and activity data from cloud stores like Snowflake or Databricks.
Why it’s interesting: As bad actors leverage AI, fraud detection needs to stay one step ahead. MouseCat helps companies identify suspicious patterns and recommend actions.
Opalite Health
What it does: An AI medical translator that helps healthcare providers communicate with non-English-speaking patients.
Why it’s interesting: Language barriers in medicine can affect outcomes. Real-time AI translation helps clinicians deliver safer, more equitable care in diverse settings.
Sequence Markets
What it does: A single trading platform for multiple markets, including crypto and prediction markets.
Why it’s interesting: Traders and institutions often face fragmented workflows. Sequence Markets unifies execution and viewing across markets for a smoother experience.
ShoFo
What it does: A custom video index billed as a “world’s video library,” designed to organize diverse datasets for AI research and discovery.
Why it’s interesting: Better indexing and search for video datasets can accelerate AI training and make it easier to find precise clips for research or creative uses.
Sonarly
What it does: Helps production software diagnose and fix issues by reducing alert noise, identifying root causes, and suggesting fixes.
Why it’s interesting: Automating incident triage and remediation speeds recovery and reduces engineer burnout — a growing priority as systems scale.
Terranox AI
What it does: Uses AI and geospatial data to locate uranium deposits in North America.
Why it’s interesting: As demand for clean energy grows, uranium remains relevant for nuclear power. Terranox applies ML to geological data to make resource discovery more efficient.
Final thoughts
YC Winter 2026 reinforced that AI remains the dominant theme, but innovation is branching into diverse verticals — from libraries and architecture to defense and language access. These 16 startups illustrate how AI and data tools are being applied to real-world problems. Keep an eye on their pitches and early traction if you’re tracking the next wave of YC-backed companies.
Want to watch their pitches? Y Combinator has posted Demo Day videos from the cohort — a great place to dig deeper into any of the companies above. 🌐