After your first AI-built app, the internet offers infinite tutorials — and almost no curated path for what to read next. You shipped a Lovable demo or a Cursor MVP. Now what? Auth? Tests? SEO? When do you stop prompting and hire a developer? This Month 2 reading list orders ten Vibe Coding Engineer posts — mixing Month 1 foundations with Month 2 depth — so you spend evenings on high-leverage skills, not random bookmark guilt. Updated November 2026.
New here? Skim the vibe coding roadmap first. If you have not built anything yet, start with 7-day plan for beginners or how to vibe code a web app, then return to this list. This page is the capstone — not step one.
Table of contents
- Who this reading list is for
- How to use this list (suggested pace)
- The 10 posts in recommended order
- Month 1 vs Month 2 clusters
- When to skip or reorder
- What to read after these ten
- FAQ
Who this reading list is for
You qualify if at least one of these is true:
- You deployed a working URL — even if it is ugly, untested, and only you have logged in
- You exported from an AI app builder to GitHub and felt simultaneously proud and terrified
- You vibe coded in Cursor or Claude Code and merged diffs you only half understood
- Real users (or a cofounder) asked for auth, billing, or “why is it slow on mobile”
This list is not for day-zero beginners picking their first tool. It is for the awkward middle: post-demo, pre-product. You know AI can build fast; you are learning what still breaks when software meets reality. The posts below address deployment, identity, quality, growth, workflow upgrades, and knowing when to hand off — in an order that minimizes rework.
| Stage | Typical feeling | Post cluster that helps |
|---|---|---|
| Just shipped URL | “It works on my machine” | Deploy, stack, roadmap |
| First real user | “Anyone can see this?” | Auth, tests, review habits |
| Adding features | “AI broke three files” | Prompt chains, MCP |
| Want traffic | “Google has not noticed me” | SEO, analytics |
| Hitting limits | “I need a real developer” | Hand-off framework |
How to use this list (suggested pace)
Read one post per evening or one per weekend deep-dive. Apply one actionable item before moving on — do not binge ten articles and ship nothing. Suggested cadence:
- Week 1 — Posts 1–3: orient, stack, harden deployment
- Week 2 — Posts 4–5: auth + tests on your live app
- Week 3 — Posts 6–7: refactor safely, then SEO basics
- Week 4 — Posts 8–10: analytics, MCP workflow, hand-off decision
Keep your first app open in a second tab. Every post assumes you are learning on a real repo, not hypotheticals. Bookmark this page; it is the Month 2 end cap that ties the calendar together.
The 10 posts in recommended order
1. Vibe Coding Roadmap — re-anchor before you sprint
Why read it now: Your first app taught you one tool’s quirks, not the full discipline. The roadmap maps phases from idea to production handoff, vocabulary (vibe engineering, thin slices, review gates), and where Month 2 tutorials sit. Five minutes here prevents choosing random next steps that fight each other — like adding Stripe before auth or SEO before a stable deploy.
Apply after reading: Label your project phase (prototype, MVP, production candidate). Pick one goal for the next two weeks that matches that phase — not three competing goals.
2. Vibe Coding Starter Stack 2026 — Cursor + Supabase + Vercel
Why read it now: First apps often accrete random dependencies — Firebase here, PlanetScale there, deploy script from a Discord tip. The starter stack documents a coherent 2026 default: IDE, database/auth, hosting, and how they connect. Even if you did not use this exact combo, understanding the reference stack makes every other tutorial easier to map to your repo.
Apply after reading: Audit your stack against the guide. Note gaps (no staging env, no migration strategy). Fix one gap — usually Preview deploy env vars or a Supabase project split for staging.
3. How to Deploy a Vibe-Coded App to Vercel
Why read it now: “I deployed once” is not the same as a reliable pipeline. This tutorial covers export paths from builders, GitHub bridge, env vars that break builds, preview URLs, custom domains, and the failure table AI codegen earns. Re-read even if you have a live URL — most first deploys skip env scoping, lockfiles, or root directory settings.
Apply after reading: Run a clean deploy from a fresh clone. Document required env vars in .env.example. Confirm Preview and Production use different secrets where appropriate.
4. How to Add Authentication to a Vibe-Coded Next.js App
Why read it now: Public demos become products when users have identity — saved data, roles, billing. AI often stubs fake login UI without sessions, RLS, or secure cookies. This post compares Clerk vs Supabase Auth with vibe-coding-friendly steps and security notes AI skips.
Apply after reading: Implement real auth on one route. Protect at least one API route or server action. Test logout and session expiry — not just the happy-path login screenshot.
5. How to Write Tests for AI-Generated Code
Why read it now: Every new prompt is a regression lottery. Tests — even a handful of Vitest specs on utils and one API route — let you prompt aggressively without silent breakage. This guide targets speed: what to test first, what to skip, mocks for Supabase and fetch, CI-friendly habits.
Apply after reading: Add five tests covering your most critical business logic. Wire npm test into your workflow before the next multi-file AI refactor.
6. Prompt Chains for Multi-File Refactors
Why read it now: Single-shot prompts collapse on features that touch routes, components, types, and DB schema together. Prompt chains break refactors into reviewable steps with checkpoints — the safe way to scale vibe coding past a todo app. Read this before your next “add dashboard” or “rename everything” session.
Apply after reading: Run one chained refactor on a branch with tests from post #5 watching. Merge only when green.
7. SEO for a Vibe-Coded App
Why read it now: AI-generated Next.js apps often ship with duplicate titles, missing metadata, client-only content Google cannot index, and sitemap gaps. SEO is not vanity — it is how solo builders get distribution without ad spend. This post covers App Router metadata, structured data basics, and traps in AI boilerplate.
Apply after reading: Fix title/description on your top three pages. Submit sitemap in Search Console after post #8.
8. Analytics and Search Console Before Launch
Why read it now: You cannot improve what you do not measure. GA4 and Search Console setup for vibe-coded sites — including SPA pitfalls, consent banners, and verifying domain ownership — belongs before you announce, not after. Pairs directly with SEO post #7.
Apply after reading: Install analytics with a privacy-respecting config. Verify Search Console property. Define one metric you check weekly (signups, activation, organic impressions).
9. What Is MCP? Model Context Protocol for AI Coding
Why read it now: After fundamentals, workflow upgrades yield the highest leverage. MCP connects Cursor and Claude to structured tools — databases, CMS APIs, design systems — so agents stop hallucinating file paths and inventing API shapes. This is the Month 2 concept post that redefines how you vibe code complex integrations.
Apply after reading: Enable one MCP server relevant to your stack (Supabase, filesystem, or CMS). Complete one task via MCP that previously required copy-paste from docs.
10. When to Stop Vibe Coding and Hand Off to a Developer
Why read it now: End on the decision framework, not denial. Some projects should stay vibe-coded forever; others need a senior engineer for compliance, performance, or architecture. Knowing the line saves money and ego. Read this before you prompt your way into a security incident or unpayable tech debt.
Apply after reading: Score your project on the post’s criteria (traffic, revenue, compliance, team size). Calendar a honest review date — 30 or 90 days — and decide: continue solo, hybrid, or hire.
Month 1 vs Month 2 clusters
The ten posts above span both months intentionally. Use this map if you want themed weekends instead of strict order:
| Cluster | Posts (by number above) | Outcome |
|---|---|---|
| Foundation | 1, 2, 3 | Stable deploy + mental model |
| Product hardening | 4, 5, 6 | Auth, tests, safe refactors |
| Growth | 7, 8 | Discoverability + measurement |
| Advanced workflow | 9 | MCP-connected agent work |
| Strategy | 10 | Build vs hire clarity |
Month 1 gave you pillars — roadmap, 7-day plan, tool reviews, workflow essays. Month 2 added tutorials (deploy, auth, tests, analytics), comparisons, prompt packs, and CI/CD. This list selects the highest overlap with “I have an app” problems rather than every calendar entry.
When to skip or reorder
- Already on Netlify, not Vercel? Still read deploy post #3 — env vars and preview patterns transfer; swap host-specific steps mentally
- Auth done on day one? Skim #4; spend time on RLS and session edge cases you may have missed
- B2B with no SEO need yet? Swap #7–8 later; do #5–6 first, then growth posts when you launch publicly
- Technical cofounder joined? Share #5, #6, and #10; they own tests, refactors, and architecture calls
- WordPress or static site, not Next.js? Auth and some SEO details differ — take principles, not every code block
Do not skip #10 because optimism feels better. The hand-off post is the guardrail that keeps vibe coding sustainable.
What to read after these ten
Once you finish the list, branch by goal:
| Goal | Next read |
|---|---|
| Automate quality gates | CI/CD on GitHub Actions |
| Review AI diffs systematically | How to review AI-generated code |
| Pick your next tool | Lovable review, Claude Code vs Cursor comparisons |
| Accessibility pass | Accessible UI prompts |
| Design handoff | Figma variables and motion token posts from Month 2 |
Revisit the full Month 2 calendar for tool reviews (Bolt, Tempo, Supabase vs Firebase) and prompt packs when your stack choices stabilize — reading every comparison before you have a repo wastes attention.
FAQ
Should I read all ten before building my second app?
No. Read 1–3, deploy cleanly, then alternate building and reading. The list orders priorities; your second app can start after post 3 if deployment was your main gap. Apply beats accumulation.
Why is this post not on the list of ten?
This page is the index — meta by design. The ten links are the curriculum; you are reading the syllabus. Bookmark it, share it with a friend who just shipped, then work through the linked posts.
How does this compare to the 7-day beginner plan?
The 7-day plan builds from zero to first app. This list assumes that milestone (or equivalent) and grows the app toward product quality. Sequential, not redundant — do beginners plan first if you have not shipped.
Can I substitute a tool review for a tutorial?
Tool reviews help you choose; tutorials help you fix your repo tonight. After your first app, favor tutorials (#3–8) over comparisons unless you are actively switching stacks. Wrong order is reading Lovable vs Cursor before you have tests.
Soft CTA: Open post #1 tonight — vibe coding roadmap — and name one skill gap your first app exposed. Read the matching entry from the list this week. Ten posts, four weeks, one app that stops feeling like a fragile demo. When you are ready to automate merges, continue to CI/CD for vibe-coded projects and never ship a red build again.
