Zeroset’s product team is building Nebula, a state layer for agents, and the foundation for enterprise automation. Every workflow an organization runs leaves a trail: decisions, actions, context earned. Today it all evaporates the moment the work is done. The state layer is the substrate that keeps it: a persistent, typed record of how work flows through an organization, so automation compounds and gets smarter every time it runs. Nebula is how that capability reaches developers. The mission is simple to say and hard to build: give every AI system a memory that grows. There is a staggering amount of work ahead, and the people who build this layer will shape what the next generation of AI products can do.
What you'll work on
- The Nebula dashboard, SDK surface, and developer tooling. Anything a customer touches.
- Interfaces for memory exploration, trace lineage, workflow visualization, and graph inspection.
- The API contracts between the platform and the people building on it.
- Sitting with users, watching them try to do real work in the product, and shipping the next iteration the same week.
- The visual and interaction language Nebula presents to developers.
You should have
- Full-stack range: modern frontend, backend APIs, data-heavy interfaces.
- Product instinct (the kind that turns capability into a thing a stranger can use without a demo).
- Shipping velocity.
Bonus
- Developer tools, observability UIs, AI product surfaces, enterprise deployments.