Back to work

Ruby AI

Live
At Grayhat

Ruby is an AI-powered B2B sales platform that acts as an autonomous deal coach — analyzing every customer interaction, updating deal strategy in real time, and generating sales content tailored to each account. Trusted by sales teams at companies like Superhuman and Nue.io. I joined as an intern and grew into a core team member, eventually owning the entire frontend of the application — from the beta UI through V1 and into V2.

Live preview
Ruby AI

The Product

Ruby gives B2B sales teams an AI deal coach that works automatically. After every customer call, Ruby deploys a set of agents that update deal strategy, surface risks, generate follow-up content, and score the deal against frameworks. It integrates with HubSpot, Gong, Zoom, Slack, and Salesforce.

The product is live, has paying enterprise customers, and has been cited by sales leaders for cutting rep ramp time in half.

My Role

I started as an intern and earned my way onto the core team. By the time I left, I was in architecture discussions, client meetings, and daily standups alongside the founders — and contributing to the backend as well as owning the frontend.

Over the course of the engagement I built:

  • The beta UI that went to first users
  • ~80% of V1, the version that brought in paying customers
  • The full V2 draft, including redesigned dashboards, analysis pages, and settings

Specific surfaces I owned: authentication and onboarding flows, dashboards, analysis and deal intelligence pages, settings and account management, loading and empty states across the app.

The Hard Problems

Payments, Subscriptions, and Refunds

The most complex feature I worked on was the billing system, Paddle integration, subscription tier management, upgrade/downgrade flows, and refund handling. This wasn't a simple "add a card" integration. It required careful state management across subscription status, grace periods, and guard rails, and had to be bulletproof since billing errors directly affect revenue and customer trust.

Document Rendering and Preview

Ruby generates rich sales content, that also includes documents like PDFs or Presentations that needed to render cleanly inside the app with real-time preview. Building a reliable document rendering system that handled different content structures, looked polished, and performed well under dynamic AI-generated content was a significant frontend challenge.