The hardest problem in robotics isn't the arm or the gripper. It's the data. Training a general-purpose robot requires a dataset that mirrors the messy, unpredictable physical world, a resource that remains scarce and expensive. Cortex AI, a three-person startup out of Y Combinator, is betting its $6 million seed round on building that dataset first [Y Combinator, 2025] [Preqin, 2025].
The data wedge
Cortex AI's stated mission is to build "the world's most diverse real-world, real-workplace, and industry-scale egocentric and robot datasets" [cortexrobot.ai]. This is a classic infrastructure-first play. Instead of starting with a robot or a specific model, the company is focusing on the foundational layer: capturing first-person video and sensor data from both humans and robots performing tasks across varied industrial and commercial environments. The premise is that whoever owns the most comprehensive, high-fidelity dataset of physical interactions will have a decisive advantage in training the next generation of embodied AI models. This approach mirrors the early days of computer vision, where ImageNet's scale and diversity became a prerequisite for progress.
The founder's pivot
Lucas Ngoo, the solo founder, brings a proven track record in scaling a consumer marketplace, but this is a sharp pivot into deep tech. As a co-founder of Carousell, he helped grow the Southeast Asian social commerce platform from an $800,000 seed round to a $35 million Series B, earning a spot on the Forbes 30 Under 30 Asia list [TechCrunch, 2013] [TechCrunch, 2016] [Forbes, 2016]. He stepped back from day-to-day operations at Carousell in early 2024, citing a new focus on AI [Yahoo Finance, 2024]. The backing from Matrix Partners and investor Taro Fukuyama suggests his operator credibility has translated into investor confidence for this ambitious, capital-intensive bet [Preqin, 2025].
| Founder | Role | Prior Experience |
|---|---|---|
| Lucas Ngoo | Founder & CEO | Co-founder of Carousell; Forbes 30 Under 30 Asia (2016) [TechCrunch, 2013] [Forbes, 2016] |
The scale challenge
For this bet to work, Cortex AI must solve two monumental engineering problems: collection and annotation. Gathering "industry-scale" egocentric data means deploying sensor suites across multiple customer sites, navigating logistics, privacy, and hardware reliability. Then, the raw video and telemetry must be annotated with high-quality labels that describe actions, object interactions, and environmental context,a process that is notoriously labor-intensive. The technical breakdown is straightforward but daunting:
- Collection Fidelity. The sensors must capture data granular enough for model training without being so intrusive they disrupt the work being recorded.
- Annotation Throughput. Manual labeling won't scale. The company will need to develop sophisticated auto-labeling pipelines, likely using foundation models, to keep pace with data ingestion.
- Dataset Bias. "Diverse" datasets must actively combat sampling bias to ensure robots trained on them can generalize beyond a narrow set of captured environments.
The sober assessment is that the wheels come off if the data collection operation cannot achieve the necessary scale and quality to be meaningfully differentiated. A dataset that is merely large, but not uniquely comprehensive or well-structured, becomes a commodity. The capital will burn quickly on hardware deployments and cloud storage bills before a single model is trained. The company's next twelve months will be a pure execution test: can it move from a compelling thesis to signed data-collection partnerships with real warehouses, factories, or logistics centers?
Sources
- [Y Combinator, 2025] Cortex AI: Large-scale real-world robot & human data for embodied AI | https://www.ycombinator.com/companies/cortex-ai
- [Preqin, 2025] Cortex AI Seed Round | https://www.preqin.com
- [cortexrobot.ai] Cortex AI, Real-World Data for Embodied AI | https://cortexrobot.ai/
- [TechCrunch, 2013] Marketplace App Carousell Raises $800K Seed Round Led By Rakuten | https://techcrunch.com/2013/11/13/marketplace-app-carousell-raises-800k-seed-round-led-by-rakuten/?_guc_consent_skip=1600855919
- [TechCrunch, 2016] Southeast Asia-based Carousell raises $35M for its social commerce app | https://techcrunch.com/2016/08/01/southeast-asia-based-carousell-raises-35m-for-its-social-commerce-app/
- [Forbes, 2016] Lucas Ngoo, Siu Rui Quek - 2016 30 Under 30 Asia: Consumer Tech | https://www.forbes.com/pictures/gegd45efke/lucas-ngoo-siu-rui-quek/
- [Yahoo Finance, 2024] Carousell co-founder Lucas Ngoo steps down, citing 'personal decision' | https://sg.finance.yahoo.com/news/carousell-co-founder-lucas-ngoo-031843387.html