Theme: The Philosophy
Length: ~600 words
Hook: Values-driven — explain the open source philosophy
We open source the infrastructure layer. We keep the intelligence layer proprietary.
Here's why.
Infrastructure is trust. If you're an enterprise and we tell you "our governance framework will keep your agents safe," you should be able to verify that claim. If the governance code is closed, you're trusting us. If it's open, you're trusting the code.
Closed-source governance is a black box. Open-source governance is an audit. Enterprises don't put black boxes in their production environments — not for security, not for compliance, and not for AI.
Infrastructure is community. No single team will think of every policy pattern, every evaluation metric, every governance scenario. By open-sourcing the infrastructure, the community contributes templates, rules, and patterns that no single company could build alone.
This is how Kubernetes won. Not because Google's code was perfect, but because the community made it theirs. The same will happen with agent governance.
Infrastructure is standards. When governance is proprietary, every vendor has their own framework. Enterprises get locked in. Interoperability dies. The industry fragments.
When governance is open, standards emerge. Organizations can move between tools. Interoperability becomes possible. The industry grows.
But intelligence is different.
The intelligence layer — the models, the fine-tuning, the proprietary data, the optimization strategies — that's where the real differentiation lives. That's where the moat is. That's what you keep.
The analogy: Google open-sourced Kubernetes but kept Search proprietary. Meta open-sourced React but kept the News Feed algorithm proprietary. The infrastructure becomes a commodity; the intelligence becomes the business.
For us, this means:
- Open source: Agent governance framework, agent DSL (language + compiler), audit tools, educational platforms. These are infrastructure — they should be shared, audited, and community-owned.
- Proprietary: Model optimization, fine-tuned models, proprietary data pipelines, enterprise integrations. These are intelligence — they're where we create unique value.
The license matters too.
We use MIT for developer-facing tools (languages, governance, audit, labs) because it's the most permissive and adoption-friendly. At our current stage — small family business, new company — the priority is adoption, not monetization. We'll consider dual-licensing or enterprise features when we have market leverage.
For products with natural paywalls (illustrated storybooks, SaaS, compliance), we use Apache 2.0 with a commercial royalty rider. Different products, different licenses, same philosophy: open the infrastructure, keep the intelligence.
The long game:
Open-sourcing infrastructure isn't charity. It's strategy. Every star on GitHub is a developer who might use our tools. Every enterprise that adopts our governance framework is a potential customer for our intelligence layer. Every contributor is a potential hire.
The infrastructure builds the audience. The intelligence builds the business.
Open source the roads. Charge for the destination.