🧠 Build Python Web Apps Without Writing Code

AI-powered development for experienced programmers

Are you an experienced developer curious about Python, web apps, or AI — but not ready to start from scratch?

I’m Tony (aka Uncle Gizmo), and I’ve built many Access and VBA systems over the years. Recently, I used large language models (LLMs) like ChatGPT, Ollama, and Gemini to help me create UberChatMaster — a full-featured, Python-based AI chat app that I built without learning Python the hard way.

Now I’m sharing how I did it — so you can do the same.

🎥 What This Course Covers:

• How I used LLMs to design, prompt, and debug a real web app
• My development process: prompt → test → refine → integrate
• How I thought like an architect, not a syntax monkey
• Tips for collaborating with AI to build tools — using your existing skills
• Practical walkthroughs for developers who think in logic

📦 What You Get with Pre-Order:

• Early access to the first video walkthroughs
• Behind-the-scenes breakdowns of how UberChatMaster works
• Prompting examples and optional downloads
• Future course updates as they are released

💸 Pre-order now for $10
Once 10 people join, I’ll publish the full series.

You’re not buying the UberChatMaster app — you’re buying my process so you can build your own.

NOTE:- This is a pre-order. Full content will be released once 10 people subscribe — expected June 2025.

✨ My story — Could be yours too.

🚀 From Chrome Extensions to UberChatMaster: How I Became an AI Application Architect Without Learning Python

By Tony (Uncle Gizmo)

For decades, I worked in Microsoft Access and VBA. It was my comfort zone — a powerful toolkit where I could move quickly, automate workflows, and solve problems efficiently. Python, JavaScript, Flask, and APIs? That was someone else’s territory.

But everything changed when I built a Chrome extension — not by learning JavaScript from scratch, but by learning to collaborate with AI.

🗣️ The Trigger: A Transcription Tool That Felt Unsafe

I’d been using a free web-based speech-to-text tool for years. It worked well, until one day I got a Google warning: “This site may send data to third parties.” That got my attention.

I didn’t want my transcriptions floating around on someone else's server. So I turned to ChatGPT and said:

“I want to build a Chrome extension that opens a page where I can speak and have my speech transcribed. The page should also let me save and clear the text — and I want the whole thing to work locally.”

A few wrong starts, some frustrating bugs, and a lot of back-and-forth later… I had my first AI-built tool:
🛠️ The Nifty Transcription Tool.

🧠 I Wasn’t Coding — I Was Orchestrating

I didn’t write a single line of JavaScript by hand. I simply:

• Explained what I wanted
• Prompted the LLM to build it
• Tested what it gave me
• Debugged with its help
• Refined, refined, refined

This was my first realization that I didn’t need to learn a new language. I already knew how to think like a developer. The AI handled the syntax — I handled the architecture.

💬 My Second Project: Saving Chat Conversations

Once I started using LLMs regularly, I hit another snag: there was no easy way to save a full ChatGPT or Grok conversation. Copy/paste was clunky and error-prone.

So I described the problem to ChatGPT and built a second extension:
📝 Chat & Grok Saver

It detects if you're on a ChatGPT or Grok page, then with one click:

• Extracts the full Chat --- ⚠️ Tip: Save regularly — huge chats might not always capture cleanly.
• Prompts for an optional file prefix
• Saves a clean .txt file to your Downloads folder, complete with model name and timestamp

Again — all done by orchestrating AI-generated code. No deep dive into Chrome APIs or content script quirks. Just guidance, testing, and correction.

🔁 Realization: These Are Stepping Stones

Each project gave me more confidence. Not in writing code — but in managing the LLM.
I wasn’t a JavaScript expert. I wasn’t a Python dev. But I was starting to act like a system architect with an AI development team.

That mindset shift changed everything.

🧱 The Bigger Project: UberChatMaster

Eventually, I wanted something more ambitious:

• A web interface to interact with multiple LLMs
• Seamless switching between cloud-based models (OpenAI, Gemini, Grok)
• Integration with local models via Ollama (e.g. Mistral)
• Persistent chat history saved in a local SQLite database
• Prompt templates, voice input, and easy chat management

That became UberChatMaster — my flagship AI tool.
Built in Python and Flask. Frontend in HTML/JS. Backend assisted by ChatGPT, Gemini, and Grok.

💡 I’m Not a Python Coder. I’m an AI Application Architect.

What I’ve built works. But I haven’t become a syntax-driven programmer.
I’ve become a strategic designer of AI-integrated tools.

I prompt. I guide. I debug.
The LLMs do the writing — I ensure the results align with my vision.

🎥 Want to Learn This Process?

I’m putting together a video series for developers who want to learn this method.
You don’t need to master Python. You need to master the collaboration loop between your logic and the LLM’s syntax.

You’ll see:

• How I built these Chrome extensions from scratch
• How UberChatMaster came together
• How you can apply the same process to your own project, in your own language

👉 Pre-order here for $10: https://gumroad.com/l/buildwithai Once 10 people sign up, I’ll launch the full series.

🎁 Free Bonus Tools

If you’re curious, you can try the very tools I built:

• Nifty Transcription Tool – Free with code NIFTY100
• Chat & Grok Saver – Free with code SAVE100

They're available via my Gumroad shop — links inside the main product page.

This is the future of coding — not just writing code, but designing tools with AI.

And if you’ve been working in Access, VBA, Excel, or any structured dev environment, you’re already halfway there.