Icamefromthesameplace.

When my roommate showed me Mr. Oklock's video last summer, I immediately told them—this was exactly how I spoke English 20+ years ago. If you compare my current English with Mr. Oklock in the video, it's significant improvement. This is that story.

Read my journey

Chapter 1

Where I Started

I still remember primary school. The English teacher asked us to pronounce vocabulary. I didn’t know how—so I asked my deskmate to write Chinese under each word so I could read Chinese instead of English.

I had poor English foundation—a small-town kid with “deaf and mute” English. No language talent. College entrance exam: 106/150. CET-4: 469. CET-6: 426. After that, I never touched English exams again.

I tried multiple times to improve my English after graduation. Failed every time.

English was the top challenge that stopped me from going abroad. When I was an undergraduate, I thought about going abroad—but when it came to my English level, I gave up the idea.

My starting point

Academic Record

OFFICIAL

College Entrance

106/150

CET-4

469

CET-6

426

2020

Chapter 2

COVID Changed Everything

Reading Mock

Practice Mock

5.5

Band Score

MeasurementDuration: 3 Hours

IELTS Official

Official Result

6.5

Band Score

MeasurementTarget Achieved

“If I couldn’t achieve this, I wouldn’t go abroad.”

It wasn’t until 2020, after COVID, that I restarted learning English. I admitted my foundation was terrible and started from the basics—vocabulary, one word at a time.

Since I used Twitter every day and already read English posts, I thought I was ready. But when I first completed a reading mock in May 2021, I scored only 5.5—and it took me 3 hours.

Reading and listening improved relatively easily. But writing and speaking? I had no idea what to do.

Since I had a full-time job, I hired personal one-on-one tutors for Speaking and Writing. It made a difference. I got some techniques.

March 25, 2022—I got my 6.5.

One of the most exciting days of my life. This was the target I set for myself: if I couldn’t achieve this, I wouldn’t go abroad.

雅思考试复盘

2023

Chapter 3

Finding Confidence

Speaking was still my top issue. In 2023, I received my school offer and decided to go. Honestly, I was still not sure...

It wasn’t until I completed my New Zealand trip that everything changed. 11 native speakers told me my English was good—even an American English teacher said it was better than most Chinese people she’d met.

I realized my English was enough for daily conversation. Then I submitted my study permit.

New Zealand trip

New Zealand, 2023

native speakers told me my English was good

Chapter 4

The GPT Era

chatgpt.com

The GPT that started it all

Joe showing the IELTS Speaking Simulator

IELTS Speaking Simulator GPT 2024

I had already used ChatGPT to create IELTS Reading and Writing feedback GPTs. But speaking was still my main issue.

November 2023—OpenAI released GPT with voice. In the shower, I had an idea: what if I could simulate an IELTS speaking exam?

I built the IELTS Speaking Simulator GPT that December and open-sourced it. I believed it was extremely helpful—and realized it would be an excellent example for AI since it has clear benchmarks for evaluation. Even hallucination is a feature, not a bug.

I used it daily, updated it, and released V3. After work, I’d run a simulation, get feedback, copy it to Notion. One issue: I rarely reviewed the feedback—I just practiced. Even on the same topic, I kept making the same mistakes. I don’t like memorizing...

I posted it on RedNote to help more people. It became one of my viral posts.

IELTS Speaking Simulator GPT open-sourced it V3

2024

Chapter 5

Canada

August 2024—I moved to Canada for my master’s degree.

Surprisingly, one of my classmates already knew my IELTS Speaking Simulator GPT. I asked him how I could improve it—I didn’t have a clear idea yet.

I wanted to build a standalone product, but I didn’t have the capacity. My thought was just to update the question bank every season.

Re:Zero − Starting Life in North America

Chapter 6

Expensive Dreams, Then Giving Up

$2.00+

Per 2-minute conversation

Built during AGI Venture Canada hackathon

“build something that belongs to yourself”

Plan: Fix 1 bug

→

Reality: 3 new bugs

November 2024—OpenAI released the real-time API. Impressive experience, just like the current product. I tried the playground and built a demo.

But it was expensive. A 2-minute conversation cost over $2.

2025 became the year of vibe coding. I gradually went from non-technical to building frontend projects with Cursor, and then Claude Code.

June 2025—Before my first trip to the Bay Area, I updated the IELTS Speaking Simulator question bank again on RedNote. Surprisingly, this post went viral—crazier than before.

The most important lesson from that trip: build something that belongs to yourself.

After returning to Ottawa, I had a plan: build a prototype using OpenAI’s real-time API. AGI Venture Canada was hosting a vibe coding hackathon—just in time!

I built a quick demo using free Anthropic credits from my Silicon Valley trip. What a coincidence! Surprisingly, it worked.

After the hackathon, I spent two more weeks polishing and launched the standalone app.

But here’s the truth: I had thought about building a real product. However, my capability—and even Sonnet 4.0—weren’t enough. I was afraid to modify features. Fixed one bug, received three more.

So I gave up. I planned to wait for more powerful AI.

the year of vibe coding AGI Venture Canada standalone app quick demo

Chapter 7

Immigration Reality Check & Mark

One of my goals in Canada was to achieve CLB 9 (IELTS 8777) to get PR after graduation. So after 11 months, I started re-practicing IELTS using my own product.

Even though I tried to use English everywhere—attending networking events mainly to practice speaking, not for networking—my speaking band score didn’t change. Still ~6.5. Rarely 7.0.

I was critical of the AI. But after checking the feedback, I realized it was valid. The issue was myself. I didn’t have the patience to re-practice the same topic multiple times. I didn’t have the courage to listen to my own voice. The issue wasn’t quantity—it was quality.

Mark

Former army officer, economist, United Nations employee. Currently retired full-time grandpa and part-time “citizen in action.”

Reporter Joe Speaking

Thank you, Mark.

Then there was Mark.

In my first month in Canada, before a networking event, I met Mark—a non-native English speaker who sounds like a native. Former army officer, economist by trade, and even United Nations employee, currently retired full-time grandpa and part-time “citizen in action” (in his own words) who is very happy to share his life and language learning experiences with everyone seeking new skill levels. Mark’s brand philosophy: “No brand, just be yourself.”

After he learned I wanted to improve my English, he started sending me voice messages—teaching me real-world communication. He gave me mini-programs like “right here right now” and encouraged me to report my day in no more than 2 minutes of audio.

At first, I just recorded some. Not consistent.

It wasn’t until last summer that I took English more seriously. I started sending my daily report to Mark every single day—and received his feedback. He even suggested I call it “Reporter Joe Speaking” as my brand, to build confidence.

I didn’t always notice my own improvement—but Mark and the college professor with impressive international ESL teaching experience, who was my mentor for a couple of weeks, both observed the small but noticeable progress. More importantly, I became more confident describing my daily life.

This is one of the reasons I built this product: to find a place to save my daily reports and get feedback.

If I hadn’t met Mark and without his encouragement, I might not have built this product—or developed this philosophy. Thank you, Mark.

At that time, I realized my backup immigration plan needed not just CLB 9, but CLB 10. In IELTS terms: 8.5 7.5 7.5 7.5. Impossible for me. I knew my capability.

So I tried other tests. CELPIP seemed better for me—more real-world conversation focus. I started preparing for CELPIP speaking and created feedback tools. This is where the current feedback system originated. This is also when I started listening to my own recordings and practicing the same topic multiple times.

joespeaking.com

Reporter Joe Speaking—every single day

2025

Chapter 8

Research Meets Reality

My legal status in Canada is still a master’s student. To graduate, I need a real-world project. I didn’t find any external projects interesting enough.

So I decided to turn the IELTS Speaking Simulator into my final research project.

Each day when I opened RedNote and received likes and thanks, it encouraged me to continue. I thought: focus on feedback, not just practice—that would be the moat. Getting fundamental theories to enhance the feedback loop would solve my own issues too.

After discussing with my professor Tony Bailetti, I confirmed the project. Thanks also to my friend Hai and AGI Venture Canada for accepting me as their client.

One shot, multiple birds.

I consider myself one of the best people to build this product: I’m an AI tinkerer, I have the demand, I understand the issue. “I’m going to build one of the best products in the category,” I told myself.

And it’s just the beginning.

Tony Bailetti Hai AGI Venture Canada scoping review results

eslr.hubeiqiao.com

My first time learning about mechanisms like Blocked vs. Interleaved Spacing and Pending vs. Elicited Feedback.

“I’m going to build one of the best products in the category”

Chapter 9

Opus 4.5 Changed Everything

I have high personal standards. The new product wouldn’t be just an IELTS Speaking Simulator—I wanted a whole end-to-end lifecycle product.

Mid-November 2025—I started building with free Claude Code credits from Anthropic. I began with daily recording, then built the fundamental infrastructure. I didn’t think I could launch something like this.

I just used the product every day to practice CELPIP speaking—testing it as I built.

Then Opus 4.5 launched.

This significantly changed my trajectory. I realized I could actually ship this.

After running out of free credits, I purchased my first $200 Claude Max subscription. Worth every penny. Without it, this product wouldn’t exist.

Worth every penny.

I knew the IELTS Speaking Simulator would be my killer feature. My original plan: integrate both OpenAI and Gemini in one week.

Reality: 5 weeks for Gemini alone.

Before release, 3 users purchased credits. Their first experience was terrible—connection disconnected after 10 minutes. I tried my best to make it stable. This is the top priority.

In summary: I built this whole product in less than 80 days by myself. It has bugs. Performance issues. But this is something I couldn’t imagine one year ago.

Thanks to this product, I registered my first company: Just Joe Technologies Inc.

I have the chance to take a product from idea to prototype to launch to more. It’s an incredible era. That’s why our tagline is: “Building What Was Impossible Yesterday.”

Just Joe Technologies Inc.technical details

Just Joe Technologies Inc.

“Building What Was Impossible Yesterday.”

“I don’t compare with others. I compare with myself.”

When I saw Mr. Oklock’s video, I didn’t feel inadequate. I felt proud. Do I have better speaking than last year? Yes, absolutely. That’s enough.

“Language is for communication, not perfection.”

My English is far from great. I’m struggling to achieve CLB 9/10 for immigration, honestly. But I realized: the core isn’t the test—it’s communication. As long as they understand your meaning, it’s fine.

As Mark reminded me: it doesn’t matter what grades or band scores you’ve achieved. What really matters is how well you can articulate your thoughts, how well you can choose the right words to express yourself, and how confident you are in that whole process. The answer to all of these? Practice, practice, and lots of practice.

“Building What Was Impossible Yesterday.”

I built this tool not for money or revenue—I want to help more people like me.When a classmate asked how to improve their English, my answer was: just speak. Don’t care about grammar. As long as you can articulate your idea, it’s OK.Yes, you can just speak. Or more broadly: take action. This app itself is an example.

The Documentation

I've been writing about this journey all along. Here's the unfiltered story.

中文雅思考试复盘 ENExpectation vs. Reality: AI Learning ENIELTS GPT Open Source ENIELTS Speaking Simulator V3 Update ENNew Zealand Trip ENRe:Zero − Starting Life in North America ENIELTS Simulator Standalone App EN2025 is the Year of Vibe Coding

中文

雅思考试复盘

A candid review of my IELTS experience and lessons learned.

Read on blog

Expectation vs. Reality: AI Learning

What I thought AI-assisted language learning would be vs what it actually is.

Read on blog

IELTS GPT Open Source

Why I decided to open source my IELTS Speaking Simulator GPT.

Read on blog

IELTS Speaking Simulator V3 Update

Major improvements and what I learned from user feedback.

Read on blog

New Zealand Trip

The trip that changed my confidence in English forever.

Read on blog

Re:Zero − Starting Life in North America

The immigrant journey and starting from scratch.

Read on blog

IELTS Simulator Standalone App

Why I built a standalone app beyond ChatGPT.

Read on blog

2025 is the Year of Vibe Coding

From non-technical to building frontend projects with AI.

Read on blog

This is a live show.

The product evolves with its users. I built it for myself—now it’s yours. We’re just getting started.

Honestly, I don’t know a lot. I learn by doing. The product has bugs. Performance issues. It’s alpha for a reason.

But each day, when I open RedNote and see the likes and thank-you messages, I’m encouraged to keep going.

Language is for communication, not perfection. As long as they understand your meaning, it’s fine.

Let’s keep building and exploring.

Try Joe Speaking Join the Community

Questions? support@joespeaking.com

— Joe
Founder, Just Joe Technologies Inc.

Chapter 6

Expensive Dreams, Then Giving Up

$2.00+

Per 2-minute conversation

Built during AGI Venture Canada hackathon

“build something that belongs to yourself”

Plan: Fix 1 bug

→

Reality: 3 new bugs

November 2024—OpenAI released the real-time API. Impressive experience, just like the current product. I tried the playground and built a demo.

But it was expensive. A 2-minute conversation cost over $2.

2025 became the year of vibe coding. I gradually went from non-technical to building frontend projects with Cursor, and then Claude Code.

June 2025—Before my first trip to the Bay Area, I updated the IELTS Speaking Simulator question bank again on RedNote. Surprisingly, this post went viral—crazier than before.

The most important lesson from that trip: build something that belongs to yourself.

After returning to Ottawa, I had a plan: build a prototype using OpenAI’s real-time API. AGI Venture Canada was hosting a vibe coding hackathon—just in time!

I built a quick demo using free Anthropic credits from my Silicon Valley trip. What a coincidence! Surprisingly, it worked.

After the hackathon, I spent two more weeks polishing and launched the standalone app.

So I gave up. I planned to wait for more powerful AI.

the year of vibe coding AGI Venture Canada standalone app quick demo

Icamefromthesameplace.

The Story

Where I Started

COVID Changed Everything

Finding Confidence

The GPT Era

Canada

Expensive Dreams, Then Giving Up

Immigration Reality Check & Mark

Research Meets Reality

Opus 4.5 Changed Everything

The Lessons