Hello, I'm

Charlie Crowley ESL Product Builder

I build apps for English learners. I taught ESL for years, saw what students actually needed, and now I'm building it. Six interconnected apps that share data so students get a more personalized learning experience.

In Production

2 Apps

AI-Powered

Learning

Multi-App

Ecosystem

From Teaching to Building

I taught English for years before getting into programming. Now I build the tools I wish I'd had in the classroom. I use AI to help with the code, which lets me ship apps fast and focus on making things that actually help students learn. Two apps live in production right now, with more launching in 2025.

What I Build

Apps that talk to each other through a shared database—so a mistake you make in one app helps personalize your practice in another. Lots of voice-based tools using speech recognition and AI. Some apps for students, some for teachers. Everything's deployed on Railway and Neon with proper security.

My Approach

I did freeCodeCamp's web design and JavaScript courses to learn the basics. Then I started using AI tools like Claude Code and Cursor to actually build stuff. I figure out what needs to happen, design the database, plan the APIs, and debug when things break. The AI just writes the code faster than I could.

My Philosophy

Programming in 2025 is more about understanding how systems fit together than memorizing syntax. You need to know what you're trying to build and why, then you can use AI to help implement it. I've shipped real apps to real users this way.

Current Focus

Scaling Firstly Academy while launching Teacher's Scribe, a B2B platform for ESL teachers to auto-transcribe lessons and generate student summaries. The ecosystem vision: students feel like "the apps know them" because mistakes in one app inform personalized practice in all others.

A Multi-App English Learning Platform

Six interconnected applications sharing one database, enabling cross-app intelligence and personalized learning experiences. From exam preparation to conversational practice, each app serves a unique purpose while contributing to the whole.

Shared Database

Neon PostgreSQL

UUID Architecture
Firstly Academy
WhatsEnglish
Teacher's Scribe
AI Voice Tutor
MemEnglish
MITEnglish
Cross-App Identity
Real-Time Sync
AI Personalization

Unified Authentication

Clerk JWT authentication across all apps — students log in once and access the entire ecosystem.

Cross-App Analytics

Mistakes from any app inform AI tutoring in others. Teachers see unified student progress.

Multiple Business Models

B2C SaaS for students, B2B plans for teachers, premium direct teaching — all in one database.

ESL Ecosystem Applications

Each application serves a unique learning need while contributing to the interconnected ecosystem.

WhatsEnglish

In Development

WhatsApp English Tutoring

AI-powered English tutoring through WhatsApp voice messages. Students practice conversational English anytime, anywhere with real-time corrections, grammar explanations, and natural conversation flow. Accessible 24/7 via the world's most familiar messaging platform.

Node.js WhatsApp API Deepgram Groq PostgreSQL

Teacher's Scribe

Private Beta

Live Lesson Transcription

Real-time speech-to-text for my 1-on-1 ESL lessons. Transcribes everything, emails summaries to students, tracks mistakes in the production database, and feeds context to Firstly Academy's AI tutor. Running locally but connected to prod. Working on DOCX exercise generation. Not ready for other teachers yet.

WhisperX Node.js PostgreSQL WebSockets

AI Voice Tutor

Production

Personalized AI Tutoring

Conversational AI tutor with access to student mistakes across all apps. Uses Teacher's Scribe transcripts for lesson-based personalization. Practice conversations based on actual class content. Live in Firstly Academy.

Deepgram Groq Llama Azure TTS WebRTC

MemEnglish

Planned

Viral Video Descriptions

Describe viral videos for vocabulary and fluency practice. Quick micro-tasks perfect for mobile learning. Viral acquisition funnel leading students into the broader ecosystem.

React Native Video AI PostgreSQL

MITEnglish

Concept

Professional Interview Simulator

Business English interview and meeting simulation for professionals. Practice presentation skills, meeting etiquette, and professional communication in realistic scenarios.

AI Simulation Speech AI Business English

What I Build With & How I Build It

I understand these technologies conceptually and direct AI to implement them. I make the architectural decisions, debug production issues, and ship real software—I just don't write every line of code myself.

System Architecture

  • Multi-app ecosystem design
  • Shared database architecture
  • Modular system design (STT → LLM → TTS)
  • Cross-app user identity (UUIDs)
  • API integration & service composition

Production Stack

  • React frontends with modern JS
  • Node.js + Express backends
  • PostgreSQL with Row-Level Security
  • Clerk authentication (JWT)
  • Railway, Vercel, Cloudflare hosting

AI Integration

  • Deepgram (Speech-to-Text)
  • Groq Llama 3.3 70B (LLM)
  • WhisperX (Local ASR)
  • Azure TTS (Text-to-Speech)
  • Real-time voice processing

What I Actually Do

  • Design database schemas (AI implements)
  • Plan API endpoints & data flows
  • Debug production errors & logs
  • Make UX & product decisions
  • Manage Git, deploys, environment vars

AI-Directed Development

  • Claude Code & Cursor for implementation
  • High-level concept → working code
  • Rapid iteration & prototyping
  • Understanding systems, not memorizing syntax
  • Weeks of work in days

EdTech Domain Expertise

  • ESL teaching experience (firsthand)
  • Cambridge FCE exam knowledge
  • Learning analytics & gamification
  • Student progress tracking
  • Teacher workflow optimization

Let's Build Something Great

Interested in collaborating, discussing EdTech innovations, or exploring the ESL ecosystem? I'd love to hear from you.

@charliecrowley
github.com/CharlesCrowley