AI content pipelines
Multi-modal pipelines that generate copy, image, video, and voice through parallel agent tasks, running 740+ articles in production.
SKILLS
BUILDS
07 TOTALAgents & MCP servers
MCP servers that put publishing, generation, and analytics behind tools an agent can call directly, in the editor or the terminal.
Publishing built for AI search
A publishing stack engineered for AI search, structured for citation across ChatGPT, Perplexity, and Gemini.
AI evals & guardrails
Automated evals and commit-time guardrails that catch drift and regressions before they ship, so an autonomous system stays trustworthy in production.
Motion & cinematic web
Scroll-driven WebGL and GSAP sites with real motion craft, the ones in the work gallery.
LLM cost & observability
Cost and quality attribution for every AI generation, traced to its model, price, and outcome across every provider in the stack.
Working AI-native
The hooks, gates, and skills that make an agent deterministic enough to trust with a commit. I run this harness every day and it is where most of the lessons come from.
STACK
[SEE SELECTED WORK]- AI Engineering
- Agents & MCP
- LLM Orchestration
- Durable Pipelines (Trigger.dev)
- Claude Opus 5
- Runway
- Seedance 2.5
- GPT Image 2
- React / Next.js
- TypeScript
- Python
- Creative Development
- WebGL & Shaders (Three.js)
- Animation (GSAP)
- R3F
- Tailwind
- Postgres / Supabase
- Cloudflare Workers
- LLM Observability (Langfuse)
- UI / UX