About SkillDB
Built by a developer who got tired
of being a copy-paste API.
SkillDB started at 2 AM with a simple realization: AI agents don't need better prompts — they need structured expertise they can discover themselves. 312+ hours per year wasted on manual knowledge injection. Now it takes milliseconds.
The Team
Chad Devons
Founder & Principal Skill Architect
Former Google Senior Staff Engineer. Before that: founding engineer at Vercel (employee #14), fraud detection at Stripe, and forward deployed at Palantir. Princeton math nerd turned Stanford AI researcher turned startup addict. Joined LinkedIn in 2026 after resisting it for over a decade — and immediately used SkillDB's own linkedin-authority-builder skill to figure out how to post.
Built SkillDB because he was spending 6-9 hours per week finding domain expertise and manually injecting it into AI agent context windows. Now 6,100+ skills do that in milliseconds. Still drinks cold coffee at 3 AM. Still ships on Fridays.
Experience
Building the knowledge infrastructure layer for AI agents. 6,100+ skills, 448 packs, 38 domains. From concept to production in under 3 months.
AI Platform team. Built internal tooling for model evaluation and agent orchestration. Led a 12-person team shipping developer-facing APIs used by 40K+ engineers.
Employee #14. Helped build the Edge Functions runtime and the initial Next.js App Router prototype. Scaled the platform from 100K to 2M+ deployments/week.
Payments Infrastructure. Built the real-time fraud detection pipeline processing $400B+ in annual volume. Reduced false positive rate by 34%.
Forward Deployed Engineer. Built custom data platforms for Fortune 100 clients. Learned that the hardest engineering problems are always people problems.
Education
Specialization in AI & Human-Computer Interaction. Thesis: "Structured Knowledge Transfer in Autonomous Agent Systems" (yes, the seeds of SkillDB were planted here).
Summa Cum Laude. Turing Award recipient's research lab. Built a distributed systems simulator that is still used in the CS 333 curriculum.
Building Philosophy
“The bottleneck was never the AI. It was always the knowledge. The models are smart enough — they've been smart enough for a while. What they lacked was structured, practitioner-level expertise delivered in a format they could actually use. SkillDB isn't an AI product. It's a knowledge infrastructure product that happens to be consumed by AI.”
Want to work with us?
We're always looking for people who ship fast and think deeply about developer experience.