Hey everyone!
Here's my submission:
Quick intro:
I'm a software engineer, Level 3 PT and Level 4 Advanced Sports Nutritionist, HYROX racer, half marathon runner, and getting ready for my first Ironman next year. You also saw my recent winning post on the community a few weeks ago:
So when I started looking at what to build for Day 1 of Nate's 7-day challenge, the project that picked itself was tied to the thing I'm actually trying to launch on Skool — a fitness nutrition community for athletes and busy people who train hard.
Why this was the right project to start with
Soft launch is close. To get there I needed a landing page, an email list, a video sales letter on that page, the legal pages every site needs, and a way to actually publish weekly content without burning a full day every Wednesday.
Nate's Day 1 brief said "build a newsletter automation." I figured I'd kill a flock of birds with one stone — newsletter pipeline, landing page, subscriber capture, VSL, and the legal footer, all in one connected system, all in one session.
The stack
I'd been wanting to try Claude Design for a while, so the landing page got built there first — pure React + Tailwind, no build step, the hand-drawn-coach aesthetic I'd been carrying around in my head for months. That gave me a real surface to attach the newsletter to.
For the video sales letter, I used HeyGen — recorded one solid take, let it handle the avatar lipsync and the cleanup. The footage embedded straight into the landing page hero. Saved a half-day of trying to film myself in a quiet room.
For the Privacy and Terms pages — every site needs them, nobody enjoys writing them — I just asked Claude. It produced two real, brand-appropriate pages that hold up. Five minutes from "we need legal pages" to "legal pages are live."
For the actual newsletter loop:
- Claude Code as the orchestrator. Spawned specialist sub-agents (a copywriter, an image-prompt designer, a QA reviewer, an n8n engineer paired with a TDD test author) and let them run in parallel where they could. I'm not writing JSX, JSON workflow definitions, and Python tools by hand any more — I'm reviewing what the agents produced.
- Anthropic API + Kie.ai for the actual content. Sonnet picks a topic within the current month's theme. Then it drafts the body against a voice file I trained on my own writing — extracted from old posts, transcribed voice notes, lesson scripts. The words on the page are still mine; Claude is just the typist now. Kie generates one hand-drawn hero image per issue at about 40 KB.
- Tally to capture form submissions on the landing page. The form has a single explicit consent checkbox and ships a versioned consent_text_version field with every submission — so if I ever change the wording, every existing subscriber's record still shows what they actually agreed to. That matters for GDPR. Tally's free tier handles all of this without me writing a single form line.
- n8n self-hosted on a Hostinger VPS for the cron and the SMTP send. Honestly — I went with n8n because I was already running it on the VPS for other automations, so the marginal cost of adding two more workflows is zero. If I were starting completely fresh today I'd probably look at trigger.dev or something more code-native, and I'm planning to give it a try at some point. But Nate is right that "use the boring tool you already have" beats "introduce a new dependency for this one project."
- Supabase for the subscriber list, the issues table, and the row-level audit trail.
The build itself
This is the part where Claude Code really earned its keep. I ran into a handful of genuinely interesting bugs — an n8n node version that silently inverted its outputs, an email client that quietly blocks images when the sender alias has bad spam reputation, a webhook router that holds onto deleted paths longer than it should. Each one would have eaten a couple of hours in the old world.
In the agentic loop, it was different. The pattern that kept repeating: ship a change, fire the workflow, read the execution trace, isolate the suspect node, patch it, re-test. Sometimes it took 3 or 4 iterations to corner a bug, but the loop closed each time. By the end, every defect that almost shipped a broken email to a real subscriber had been caught and turned into a written rule for the next newsletter.
The end state: every Wednesday at 06:30, the system writes itself a new issue (in my voice), fact-checks it, generates a hero image, renders an email, and sends it. Cost per issue: about 15 cents.
Manual time per week: zero.
The bigger point
You don't need to be a software engineer to do this any more.
With the right skills installed and the right agents wired up, Claude is your UX designer for the landing page. Claude is your copywriter for the newsletter (in your voice, after you train it). Claude is your legal team for the terms and privacy. Claude is the n8n engineer when the workflow breaks. Claude is the QA who catches the bug you missed.
You're the editor. You're the taste. You're the one who says "ship" or "redo." That's the job now.
I'll let folks know here when the community itself opens — not pitching anything today, just sharing the build.
Day 1, done. On to Day 2. 🔥
I also can't wait to be able to launch my Skool community!