Introducing Aonyx
We are building an AI-native development platform that helps teams ship faster and reach new platforms ahead of competitors.
AI is fundamentally changing how software is built and consumed. Teams that master AI-assisted development gain exponential advantages, while products are increasingly accessed through agents and new platforms rather than traditional interfaces. This transformation creates winner-take-most dynamics—and most development tools weren't designed for this world.
That's why we're building Aonyx: an AI-native development platform that helps teams ship faster and reach new platforms ahead of competitors. We provide an opinionated framework, integrated cloud infrastructure, and AI tools with context that generic solutions simply can't access—production metrics, error patterns, and infrastructure data. When your AI understands not just your codebase but how your application actually runs, it can generate fundamentally better code.
We're Jan David Nose and Martijn Arts. JD previously led infrastructure for the Rust Foundation and is still a team lead in the Rust Project. Martijn was the first engineer at Headroom and founded RCKIVE. We both spent time at Palantir, where we learned to obsess over customer delivery and execution. We're betting that Rust's compile-time guarantees and performance characteristics make it the ideal foundation for reliable AI-assisted development.
The problem we're solving is threefold. First, AI code generation is unreliable without strong architectural foundations—teams either move fast and break things, or move slowly and lose competitive ground. Second, implementing best practices (security, observability, compliance) remains expensive and time-consuming. Third, products need entirely new architectures for agent-first consumption, but most teams lack the resources to rebuild while maintaining their core business.
Our solution is vertical integration. By controlling the framework, platform, and AI tooling as one coherent system, we create capabilities that separate tools cannot match. Our AI doesn't just know your code—it understands your production environment, learns from usage patterns, and suggests improvements based on real behavior. The framework codifies best practices so teams ship features instead of infrastructure. And because we're building for new platforms from day one, apps built on Aonyx are ready for MCP servers, agent integrations, and whatever comes next.
We're currently building the core foundation of the open source framework, targeting early 2026 for its first public release. We're taking a measured approach: validate our architectural decisions with real code, iterate based on what we learn, then expand.
If you want to stay updated, subscribe to our blog for deep dives on AI-native architecture, building in Rust, and the challenges of vertical integration. Follow us on Bluesky and Mastodon for shorter updates. We're building Aonyx in the open and learning as we go.