// principal solutions engineer @ cloudflare
My team of agents and I brainstorm, spec, and build - often while I sleep.
Principal Solutions Engineer at Cloudflare during the day and shipping AI products by running agent fleets overnight. This is the way.
Let's Go Christo is the platform where I will share what I'm building, what I'm learning, and where it's all going. LET'S GO!
I built Daytripper and RepoFinder entirely with OpenAI Codex, then turned my Claude Code tooling loose on the results. The QA and design findings are numbered in the git history, one regression test cites the report that found the bug, and both products came out better than either agent would have produced alone.
Anthropic ships cookbooks for financial workflows. I did not see one for the problem I actually have: taking over a codebase with no docs, no tests, and no author. So I wrote it. This is the whole thing, top to bottom - the recipe, the building blocks (commands vs skills), the guardrails (ask vs enforce, hooks and all), and why Claude Code is the tool for a regulated fintech with three teams to serve.
We shipped RepoRadar in four hours at the AI Tinkerers SF Generative UI Global Hackathon: a radar that ranks trending GitHub repos, then an agent deploys a working micro-app for any of them at its own subdomain. Here's what we built, how it works, and how to grab it.
RepoFinder turns a project and a goal into a short list of open source tools that genuinely fit, then lets you interrogate every pick. Built entirely with OpenAI Codex on the Responses API: gpt-5.6-luna for extraction, gpt-5.6-terra for ranking and chat, strict JSON Schema outputs, a labeled fallback, a remote MCP tool, and 15 checked-in lessons on how Codex actually works.
I learn AI coding agents by shipping with them. Daytripper is the first app I built to learn OpenAI Codex: a travel planner built in one overnight session, deliberately zero LLM calls at runtime, with tests covering all 1,296 possible preference combinations, deployed to Cloudflare Workers. Here is what the build taught me about how Codex actually works in August 2026.
Rocket Mortgage put Claude Code in front of the engineers who own their mortgage origination platform. A migration that had dragged for nearly 3 years shipped in 4 months, senior engineer throughput went up 145%, and client-impacting incidents after the cutover: zero.
Turn a project and a goal into a short list of open source tools that genuinely fit - then interrogate every pick. Built entirely with OpenAI Codex on the OpenAI Responses API: gpt-5.6-luna extracts, gpt-5.6-terra ranks and explains, strict JSON Schema outputs, a labeled GitHub fallback, a remote MCP tool, 15 checked-in lessons on how Codex works, and a three-tier eval harness. One Cloudflare Worker, D1, plain HTML/CSS/JS.
A travel planner built in one overnight session to learn OpenAI Codex. Pick Lisbon, Tokyo, or Mexico City, set budget, time, vibe, pace, and interests, and get one thoughtful day plan with an honest cost estimate and a "why this fits" explanation. Deliberately zero LLM calls at runtime - a deterministic scoring engine, with tests that run all 1,296 possible preference combinations and check every plan is chronological, in the right city, and under budget. Next.js 16 on Cloudflare Workers via vinext.
Open-source Claude Code kit that decodes a codebase you inherited. Point it at any repo and one command produces full documentation with diagrams, an executable test suite, prioritized recommendations split for three teams (engineering, SRE, data science), and a deployed review portal you can chat with. About 30 minutes, about $30 in tokens - versus a multi-week, five-figure engagement.
Launched July 2026, the disciplined follow-up to RepoRadar. Point it at your GitHub repo and a goal ("auth solution", "background jobs") and it returns open-source repos that genuinely complement yours, with a reason each one fits and ratings for ease and impact. Built in the open with Claude Code, with every major step written up in a 23-lesson book: plan mode, CLAUDE.md, custom skills, hooks, model tiering (Haiku extracts, Sonnet reasons), and one engine serving both a Web API and an MCP server.