Forem: Chad Dyar

My Stack for Building and Monitoring 6 Web Apps as a Solo Developer

Chad Dyar — Sat, 28 Mar 2026 04:40:10 +0000

I run six web apps as a solo developer with a day job. Here is the exact stack and how the pieces fit together.

Hosting and deployment: five apps are on Lovable (one-click deploy from their builder), one (ContentForge) is on Vercel. All use custom domains through Cloudflare for DNS and SSL.

Database: Supabase (Postgres) for all six apps plus a central "command" project that holds cross-app analytics, book sales data, and the marketing engine.

Payments: Stripe with tiered pricing on every app. Free tier, mid tier ($3.99-4.99/mo), premium tier ($7.99-9.99/mo). No paying customers yet, but the infrastructure is ready.

Analytics: PostHog for product analytics across all six apps. One project, events tagged by app name. This lets me compare engagement across apps in a single dashboard.

Error tracking: Sentry with one org and separate projects per app.

Uptime: BetterStack monitors all six domains.

Email: Resend for transactional email.

The key architectural decision: one central Supabase project acts as the hub. It connects to each app's project through Supabase edge functions. This gives me a single place to query across all six apps, run marketing automation, and track portfolio-level metrics without logging into six different dashboards.

Current numbers are small (364 total pageviews last month across all six apps), but the monitoring infrastructure scales. Every tool on this list either has a generous free tier or has a free tier that covers my current scale.

The apps: contentforgehq.com, momentumfit.app, thehomegrown.app, getpillpal.app, pawformance.app, palettepro.design.

Code and details at chadtdyar.com.

The Compounding Gap: What Happens When AI Stops Resetting Every Session

Chad Dyar — Sat, 21 Mar 2026 17:22:07 +0000

About a year ago I stopped treating AI as a tool I use and started treating it as infrastructure I build on.

The difference sounds academic until you live inside it.

What the old model looked like

Open a new chat. Explain who I am, what I'm working on, what my standards are. Get a draft. Close the tab. Next day, repeat. Every interaction was a cold start. The AI was a temp worker who showed up fresh every morning with no memory of yesterday.

What the new model looks like

Context persists. My AI systems know my writing voice, my active projects, my preferences for how I structure arguments and present data. When I sit down to work, I'm not onboarding a new assistant. I'm picking up where I left off.

The shift didn't require exotic tooling. It required asking a different question. Instead of "can AI help with this task" I started asking "what would this workflow look like if AI were present at every layer?"

Where the gap is now

Most people I talk to are still in the old model. New chat, fresh context, isolated sessions. They're "using AI" the same way they were a year ago.

The people who shifted to persistent systems six to twelve months ago are in a different category. Their outputs are compounding because every interaction builds on the last. The reset tax they used to pay every morning is gone.

From the outside, both groups look the same. Both "use AI." But the outputs are diverging in ways that are increasingly hard to ignore.

The part nobody says out loud

This gap will keep widening. The barrier to building persistent context is dropping every month. Memory systems, custom instructions, persistent agents, these are all getting more accessible. The people who start building now will catch up faster than early adopters did. The people who wait will be further behind than they expect.

What I'd tell someone starting today: stop asking AI to help with tasks. Start asking what your entire workflow would look like if context never reset.

How a 15-Year Opera Career Became a Sales Training Framework (And How I'm Shipping It)

Chad Dyar — Thu, 12 Mar 2026 12:53:45 +0000

I did not expect my background in professional opera to become the core IP of a sales training book. But here we are.

The through-line is ensemble performance. For fifteen years I sang with professional opera companies. The thing that separated great ensembles from competent ones had nothing to do with individual skill. It was whether the performers were building one shared story or several individual stories that happened to share a stage.

When I moved into sales enablement, I kept watching the same pattern on account team calls. Skilled individual sellers showing up on the same Zoom window and telling different stories. The prospect could feel it. The deals slowed or died in ways that had nothing to do with the product or the price.

The framework I built from this is in a book called Think on Your Feet, Land on Your Numbers. The core principle is "yes, and" from improv: you build on what your scene partner gives you instead of redirecting or correcting. Applied to sales, it means your team picks up the baton from each other on calls and tells one coherent story the prospect can follow.

How I built the training module

Three components, in the order teams should learn them.

Simulation first. Before anyone reads anything, run call simulations where the only rule is that each person must acknowledge what the previous person said before adding their own content. "Building on what [teammate] just said..." is a legitimate bridge. Do this twice and the concept clicks faster than any written explanation.

Call analysis second. Pull recordings from real multi-person calls. Identify every moment where a team member failed to build on what the previous person offered. Call them "thread breaks." Count them per call. This creates a measurable baseline and surfaces specific moments to debrief.

Pre-call brief third. Four minutes, three questions, one shared destination before anyone dials in. This is the structural fix that makes the listening habit stick.

How I'm distributing it

ContentForge, which I also built and run. One core idea from the book goes in. Platform-specific content for eleven channels comes out. The blog post is the long-form anchor. Everything else adapts to format. I'm testing whether leading with the opera backstory outperforms leading with the sales principle. My bet is the personal story pulls harder. Early data this week will tell.

The lesson I'd give builders creating training content: experiential first, conceptual second. Always. The teams that adopted the framework fastest did the simulations before they read a single page of the book.

The Distribution Layer Nobody Sees

Chad Dyar — Thu, 12 Mar 2026 00:50:02 +0000

I'm running 6 AI apps and 16 books as a one-person operation alongside a full-time job. Here's the distribution architecture I shipped in 48 hours.

Each app gets a free guide targeting its audience, a 3-email Kit nurture sequence with segment-specific tags, a landing page with embedded form, and a book that bridges guide to app. Guide, book, app. Six times.

I wrote a /health/forms endpoint that reads each landing page's HTML, extracts the data-uid, and confirms the Kit embed matches. Third-party form embeds break without telling you. I've lost leads to that before.

Stack: static HTML on Express. Kit for email with segment tags. UTM pattern: utm_source=lead_magnet&utm_medium=email&utm_campaign={segment}.

PillPal has 19 users, 4 paying. Other five are starting from near-zero. 90-day sprint to $1K MRR combined, day 3.

PillPal: Building a Medication Tracker That Actually Helps

Chad Dyar — Sun, 08 Mar 2026 12:38:29 +0000

Missing a medication dose might not seem like a big deal. But when you're managing multiple prescriptions, it adds up fast.

Why I Built PillPal

After a health scare, I found myself juggling multiple medications with no good system. Sticky notes failed. Phone alarms were too generic. Existing apps were bloated.

So I built PillPal — a simple, focused medication tracker.

Core Features

Smart Reminders — Context-aware notifications that actually help
Refill Tracking — Know when you're running low before it's too late
Medication History — See your adherence patterns over time
Simple UI — No unnecessary complexity

Tech Decisions

Built with React, designed mobile-first. Every feature exists because I needed it personally.

The biggest lesson: healthcare tools need to be simple above all else. If it takes more than 5 seconds to log a dose, people won't use it.

Try It

PillPal is free at pillpal.app. Feedback welcome.

More about my projects at chadtdyar.com

I Built 6 Web Apps as a Solo Dev — Here's What I Learned

Chad Dyar — Sun, 08 Mar 2026 12:34:49 +0000

A health scare in my 30s changed everything. I stopped waiting for "someday" and started building tools I actually needed.

What I Built

Momentum — A habit and goal tracker for people who want clarity without gamification noise. Track what matters, stay accountable.

PillPal — Medication management with smart reminders and refill tracking. Born from missing too many doses.

The Deliberate Breath — A book about slowing down and breathing with intention.

The Stack

All apps built with React, deployed on modern infrastructure, designed mobile-first. No VC funding. No team. Just personal need turned into shipped products.

Key Lessons

Build from pain points you actually have
Ship fast, iterate in public
The story behind the product matters more than features
People connect with WHY you built it

All live at chadtdyar.com

Would love feedback from the community. What are you building from personal need?

I Built 6 AI-Powered Web Apps While Working Full-Time — Here's the Stack and What I Learned

Chad Dyar — Sat, 07 Mar 2026 13:11:07 +0000

Over the past 18 months, I shipped 6 AI-powered web applications while holding down a full-time Director-level role in sales enablement. Here's a quick overview of each app, the tech stack, and what I learned along the way.

The Apps

1. ContentForge HQ

Paste one blog post, get platform-native content for LinkedIn, X, Instagram, email newsletters, and more. Built for creators tired of manually rewriting the same idea 8 different ways.

Try it: contentforgehq.com

2. PawFormance

AI-powered dog wellness tracker. Log meals, walks, vet visits, and medications. The AI spots trends and gives personalized recommendations.

Try it: pawformance.com

3. Momentum

Fitness and nutrition tracker with AI-driven insights. Set goals, log workouts, track macros.

4. PillPal

Medication and supplement tracker with smart reminders.

5. Palette Pro

AI color palette generator for designers.

6. HomeGrown

Garden planning assistant with AI recommendations.

The Stack

Frontend: Lovable (React-based)
Backend: Replit
AI: Claude API (Anthropic)
Payments: Stripe
Hosting: Netlify + Replit
Database: Neon (PostgreSQL)

Key Takeaways

Ship fast, iterate faster. Every app launched with a minimal feature set.
AI as a co-builder. Claude helped write code, debug, and draft copy.
Free tiers matter. Removes friction and builds trust.
Side projects compound. Each app made the next one faster.

If you're thinking about building with AI, just start. The tools are better than ever.

I'm Chad T. Dyar — author of 15 books, executive coach, and AI strategist. chadtdyar.com