Ryan Alexander Black

AI Solutions Architect — in the making

I architect whole systems — and build them, end to end.

Business-strong builder turning ideas into working AI software. I take products from idea to live — designing how every piece flows together, then shipping it. Below are selected builds, each with the problem it solved and the stack behind it.

Selected work

Builds & case studies

Live

OCLA

A complete app that takes a fitness coach from zero to a launched business.

Software that takes a fitness coach from zero to a launched, running business — four guided modules plus the tools to operate it: content and ad planners, an AI assistant that drafts replies to clients, a simple client tracker (CRM), and a website builder. Built solo over ~300 hours.

Under the hood

  • Automated trial-to-paid flow — sign-ups and billing handled with no manual steps
  • AI hardened to stay accurate in production (it won't drift or invent its own data)
  • Row-level security — every user only ever sees their own data
  • Next.js
  • Supabase
  • Claude API
  • Vercel
Coming soon
Shipped

Newsletter Automation

"Signal over Noise" — give it a topic, get a finished newsletter.

Give it a topic and it produces a finished newsletter — researched with real sources, written in a set brand voice, with accurate charts and a header image, then checked and sent. The card below links to a real issue it produced.

Under the hood

  • Charts are drawn in code, not AI-generated — so the numbers are always exact
  • Every issue is checked by rendering it to an image before it sends
  • Each part (research, images, sending) is swappable without rebuilding the rest
  • Python
  • Perplexity
  • Claude
Read case study →
Live

TFS Creative Suite

One place to generate images, video, voice and music — saved to the cloud.

A single toolset for generating images, video, voice and music from top AI models, with every result saved to the cloud automatically. I built the connector that wires those models into the tools I use day to day — plus an automated video editor that turned clips into finished Reels for live ad campaigns.

Under the hood

  • A custom connector (MCP server) links multiple AI media models into one toolset
  • Generated assets save to permanent cloud storage automatically
  • An automated editor (Remotion) assembled clips into finished, ad-ready Reels
  • MCP
  • KIE.ai
  • Remotion
Coming soon
Planned

Challenge — Day 2

Day 2 of the 7-day AI build challenge.

One build a day, shipped publicly. Coming as the challenge runs.

  • TBD
Coming soon
Planned

Challenge — Day 3

Day 3 of the 7-day AI build challenge.

One build a day, shipped publicly. Coming as the challenge runs.

  • TBD
Coming soon
Planned

Challenge — Day 4

Day 4 of the 7-day AI build challenge.

One build a day, shipped publicly. Coming as the challenge runs.

  • TBD
Coming soon
Planned

Challenge — Day 5

Day 5 of the 7-day AI build challenge.

One build a day, shipped publicly. Coming as the challenge runs.

  • TBD
Coming soon
Planned

Challenge — Day 6

Day 6 of the 7-day AI build challenge.

One build a day, shipped publicly. Coming as the challenge runs.

  • TBD
Coming soon
Planned

Challenge — Day 7

Day 7 of the 7-day AI build challenge.

One build a day, shipped publicly. Coming as the challenge runs.

  • TBD
Coming soon
Planned

Marathon Training Plan

An adaptive training-plan builder (scope TBC).

A personal build in the running domain — concept and scope being defined.

  • TBD
Coming soon