throxy is a small applied research lab working on the future of sales. We build AI agents that generate pipeline for our customers, and we sell the outcome — pipeline that closes revenue — rather than a pile of tools. This blog is where the engineering team writes about how that actually gets built.
Expect posts about agent architecture, retrieval, data systems, evals, and the unglamorous plumbing that makes any of it reliable. Less launch-copy, more load-bearing detail.
A quick note on format: code blocks here are first-class. They come with a copy button, optional file titles, and proper highlighting — because half of what we want to share is the code.
export const hello = (name: string): string => `Hi ${name}, welcome aboard.`;This post is written in MDX, so it can pull in real components from the site. The
call-to-action below is the exact same <HireCTA /> component the rest of the blog
uses — dropped straight into the content:
If building this kind of thing sounds like your idea of a good time, the role is open and we’d love to hear from you.