Liquid Glass Lower Thirds — 10 Remotion Titles (MP4 or Transparent)
10 designs · Remotion · 16:9
Predict which students will drop out, see why, send the fix. Full Next.js 16 source.
Commercial license · GST calculated at checkout
Claude Code
Model: Claude Opus 5.5
Built and documented with Claude Code on Claude Opus 5.5.
AGENTS.md and docs/ai/ let agents change it without scanning the repo.
LearnPulse is a complete AI student-retention app for online courses. It scores every student's dropout risk from 0 to 100, names the root cause and drafts the re-engagement email, so course creators step in before a student is lost. It is a finished product, not a UI kit: a risk engine, student analytics, an AI assistant grounded in each student's data, an intervention queue and course health.
Underneath is a production platform: Better Auth with passkeys, Google, magic links and email codes; plans and prices you set in the admin, synced to Stripe as subscriptions and lifetime plans; an admin with users, plans and a blog; email through Resend, Mailgun or SMTP; S3 or R2 storage; database-backed rate limits, security headers and environment checks at boot. Light, dark and mobile, with ⌘K search and a docked AI panel.
Run it locally in minutes with Docker, two demo accounts and 40 seeded students. AI Brain works without a key in demo mode, or live with an Anky AI API key (OpenAI-compatible). AGENTS.md, CLAUDE.md, Cursor and Windsurf rules and a codebase map let Claude Code, Codex or Cursor rebrand it and add features with you.
What you are buying: The full Next.js 16 source of LearnPulse: the retention app, the risk engine, the AI Brain assistant, the admin, the database schema and demo seed, and the docs. Nothing else to buy to run it; deploy it on Vercel or any Node host with PostgreSQL.
Third-party costs: LearnPulse can connect to Stripe, an email provider (Resend, Mailgun or SMTP), S3 or R2 storage and the Anky AI API. Each bills you directly for your usage and stays off until you add its key; none is needed to try the app.
LearnPulse · AI student retention
A risk score for every student, the reason behind it and a ready-to-send email, so course creators step in before a student is lost.

0–100
Risk score per student
4
Risk levels
5
Sign-in methods
10
Email templates
1
One-time payment
Everything included
An explainable 0–100 score from inactivity, engagement change, stalled lessons, video time and schedule, with the root cause named.
A chat docked on every page that answers from the student's real data: explain the risk, predict the outcome, draft the email.
Drafted re-engagement emails you edit, approve and send through your provider. Nothing goes out on its own.
Learning activity, predicted outcome, habits, devices and a full activity timeline for every student.
Better Auth with passkeys, Google, magic links and codes, plus rate limits, security headers and workspace isolation.
Plans and prices edited in the admin and synced to Stripe, subscriptions and lifetime plans, a customer portal, and an admin for users and a blog.
How it works
Write ActivityEvent rows from your LMS webhooks or an import job: logins, lessons, videos, quizzes. The seed gives you 40 students to start.
npm run db:seed
Home and Risk Alerts rank every enrollment by risk, with the primary cause and a 48-hour churn warning.

Open a student and ask. Answers are grounded in that student's analytics, signals and notes.

Approve the drafted email in the AI Agent queue and send it through your provider.

Who it's for
Course creators and coaches
See every morning which students are slipping and send a personal nudge in two clicks.
Cohorts and bootcamps
Track a whole cohort and spot students stuck on the same lesson.
Memberships
Catch disengagement before it becomes a cancelled renewal.
Edtech startups
Launch a branded retention product with workspaces, billing and admin built in.
Customer onboarding teams
Treat onboarding tracks as courses and find accounts that stall.
Agencies
Resell a white-label retention dashboard to course-creator clients.
Make it yours
The repo ships with AGENTS.md, CLAUDE.md and a codebase map. Open it in Claude Code, Codex or Cursor and ask in plain words:
“Rebrand it as FocusTrack with an emerald accent and my logo.”
“Connect my Teachable webhooks and write ActivityEvent rows for lesson completions.”
“Add a nightly cron that re-analyzes every student and emails me the critical ones.”
“Add a Cohorts page that groups students by start date.”
Everything you need to evaluate the stack at a glance.
Ships with AGENTS.md, CLAUDE.md, GEMINI.md and .cursor/rules.
$ npm install$ cp .env.example .env$ docker compose up -d$ npm run db:migrate$ npm run db:seed$ npm run dev
Open the folder in Claude Code, Codex or Cursor and ask it to rebrand, add a feature or deploy — the instruction files explain the architecture and conventions.
10 designs · Remotion · 16:9
Next.js 16 · TS · Claude Code-ready
Next.js 16 · TS · Claude Code-ready
Next.js 16 · TS · Claude Code-ready