Investor brief
The runtime AI agents will run on.
AI agents need a runtime — and whoever owns it owns the margin. Sagewai is the open-source one: every secret scoped, every dollar tracked, every model yours to own.
Why now
Every wave of compute creates a runtime layer where the value accrues — operating systems on PCs, container orchestration in the cloud. AI agents are the next wave, and that layer is being built right now.
The frontier-model providers will not own it. Their incentive is to lock customers into hosted endpoints; the customer's incentive is to stay portable, observable, and in control of cost. An open, self-hostable agent runtime is what closes that gap.
Where we are
We optimise for proof over promises. What is true today, and independently verifiable:
- Open source under AGPL-3.0 — self-hostable end to end, running in production today.
- 50+ runnable examples in the repository, each one a reproducible demonstration.
- Cross-tenant isolation is CI-gated: the build fails if a worker can claim another tenant’s task.
- Customer credentials never reach the worker host — enforced by a CI gate, not a policy.
What compounds
A customer can start on a frontier model and, from the runs captured along the way, fine-tune a smaller model that takes over more of their traffic over time. Marginal cost trends down — and the incentive to leave trends down with it.
That is a different shape of defensibility than vendor lock-in. The switching cost isn't something we impose; it's the savings the customer built by staying.
The rest is a conversation
Go-to-market, the competitive analysis, the commercial pipeline, design partners, and the strategic roadmap are shared directly under NDA — not posted on a public page.
If the thesis lands, the next step is a call.
hello@sagewai.aiPut "investor inquiry" in the subject so it routes correctly.