Production AI systems,
built by one engineer.
I am James Phoenix. I build AI-powered SaaS products from the ground up, write about how they really get shipped, and teach the people building the next generation of them.
I build AI-powered SaaS products from the ground up. My most recent project is AI Rank Tracker, a production system with ~350K LOC that monitors brand visibility across ChatGPT, Claude, and Gemini.
Along the way, I wrote an O'Reilly book on prompt engineering and have taught 306K+ developers taught through Udemy courses and 45+ bootcamps at General Assembly.
The through-line in all of it is the same. I care about systems that work in production, not demos that break under load. That means observability, reliability, and performance at scale, baked in from the first commit rather than bolted on after launch.
I am based in London, UK, and I work across the full stack. Python backends, data pipelines, React front-ends, and the LLM infrastructure that ties them together.
Numbers That Tell the Story
Real metrics from real systems
What I'm Building Now
Three things keep me busy. A product, a book, and a curriculum.
AI Rank Tracker
Tracks brand visibility across ChatGPT, Claude, Gemini, and Perplexity. Currently processing 5K queries daily with capacity to scale to 100K. Built with a 3-stage pipeline handling complex orchestration and rate limiting.
Visit BrandVisibility.ai →O'Reilly Book
"Prompt Engineering for Generative AI" started as my notes for teaching LLMs. 5,000+ copies sold with a 4.4-star rating. It covers practical techniques that work in production, not just demos.
View on Amazon →General Assembly
Five years and 45+ bootcamps have taught me one thing: everyone can code, they just need the right guide. I created GA's first Applied AI Engineering curriculum because the industry needed practitioners, not just theorists.
5+ years shaping the next generationBattle-tested stack
From frontend to infrastructure, every tool chosen for production reliability. These aren't just technologies I know. They're technologies I've deployed, scaled, and debugged at 3am.
- ✓Full-stack TypeScriptEnd-to-end type safety from React to API
- ✓Production observabilityTraces, metrics, and logs with OpenTelemetry
- ✓Infrastructure as codeReproducible deployments with Terraform
TypeScript
React
Next.js
Python
Node.js
PostgreSQL
Redis
Docker
GCP
Terraform
OpenTelemetry
Prometheus
Grafana
Effect.ts
Temporal
ESLintReady to build something great?
Production-grade software and data engineering. Let's discuss your project.