Blog
Notes on routing, governance, and proof
Short posts on what we think the LLM gateway layer should do, and why we built RemKey the way we did.
Stripe bought the meter. Someone still has to keep the ledger.
Stripe is paying $7.5 billion for OpenRouter, months after Palo Alto bought Portkey. Two acquisitions in one quarter say the layer that decides how to spend AI is now strategic. But notice what neither buyer bought: the evidence.
The moat moved upward. Here's the layer it moved to.
When an open model ties the frontier at a third of the price, the advantage stops being which model you own and becomes the system that decides how, when, and where to use each one, with the evidence to prove it. That system has a shape, and most teams are describing it without building it.
I run an AI governance company and I still said yes to everything
A founder confession: this week I botched my own env variables, approved every tool prompt an agent showed me, and lost an evening reverse-engineering token thresholds in a compiled binary. I am the customer. Here's what that taught me about what developers will actually do.
Seven million apps, zero audit trails
Vibe coding is the fastest software wave in history: millions of production apps built by people who have never heard of prompt injection. This isn't a complaint about the wave. It's a description of what the wave leaves behind.
Hassabis wants a FINRA for AI. Banks already have FINRA.
DeepMind's CEO just proposed a financial-services-style oversight body for frontier AI. The essay is about model labs, but the analogy he chose tells you where the evidence requirements land next: on the firms that use the models.
A gateway is a ledger, not a router
Model catalogs are the commoditizing axis of this market. The durable asset is the record of what happened, and that record is only trustworthy, and only compounding, if the system that made the decision is the one that wrote it down.
Keyword routing vs classifier routing, and why the afternoon-script approach stalls
Everyone is converging on the same pattern: expensive model for the hard part, cheap model for the rest. The naive way to automate that decision caps out fast. Here's why, and what actually works.
Why we verify every downroute instead of trusting the classifier
A router can tell you it sent a request to a cheaper model. It can't tell you the answer held up. Here's the mechanism we built so that claim comes with proof instead of a shrug.
Routing isn't governance, and the gap is the opening
OpenRouter, LiteLLM, and Portkey all solve pieces of the LLM gateway problem. Here's the piece none of them solve, and why it's the one a security team actually cares about.