You’ve probably seen hundreds of public datasets. Most end up in abandoned notebooks.

This one — sugar consumption across decades — didn’t feel any different.

Until I had an idea.

What If Two AI Models Built It Together?

Then the idea hit me — what if I didn’t build the dashboard alone?

What if I asked Claude 3.7 Sonnet, the poetic visual thinker, to design the look and feel?
What if I paired it with ChatGPT o1, my logical strategist, to define what really mattered?

Suddenly, this wasn’t just data visualization.
It was an AI collaboration.
Two models.
Two styles.
One goal: turn decades of sugar data into something humans can actually feel.

Global Sugar Consumption Trends (1960–2023)

Time to meet. Let’s meet with the data.

To do that, I am going to use Randy from our platform. Randy automates data analysis.

SS of the output

After a few iterations, it shows the magic.

SS of the output

Good. Now we have an idea of what we’ll deal with. To have an idea about what we are going to create with Claude, we’ll talk with ChatGPT o1.

But before that, let’s ask Randy to summarize the data for us so that we can paste the info into o1.

GPT o1

Now that Randy did the heavy lifting on the data side, it’s time to bring in the thinker: GPT o1.

While Claude handles how things look, GPT o1 decides what things mean.

• Which insights matter?
• What should we highlight?
• How do we turn rows into understanding?

I didn’t just want one dashboard.
I wanted options — and a clear winner.

So I asked GPT o1:

Based on this dataset above, give me 8 different Streamlit dashboard ideas I could build. Each should focus on a unique angle or insight. Then compare each idea and pick the winner — the one that’s most interesting, helpful and creative for Data Science and AI.

Here are our alternatives.

Ideas

Which one is the winner? Let’s see.

Now It’s Claude’s Turn — Time to Bring the Vision to Life

With the winning idea locked in — the AI-Forecasting & What-If Simulator — it was time for Claude 3.7 Sonnet to step in.

While GPT o1 was all about logic, data, and comparison, Claude brought something different:
Design intuition. Visual hierarchy.

Human feel.

So I prompted Claude:

Design a clean, modern Streamlit dashboard for sugar consumption forecasting. It should include time series prediction, what-if sliders (like tax or policy changes), and visualizations that tell a story — not just show numbers.

After a bit of back and forth and debugging, now we have a dashboard.

Go to your working directory. Install streamlit.

pip install streamlit

Next, run.

streamlitrun sugar_consumption.py

And voila! We have a dashboard, let’s analyze it.

Data Overview

Here we saw an overview. And yes, you can customize.

Consumption Forecast

Here you can generate a forecast with the Prophet model, Machine learning with a single click.

Also, you can switch to the Random Forest.

What-If Simulator

You can run a simulation from here. Let’s see the inputs.

Here is the output.

Health Impacts

Here we saw health impacts. 

More graphs.

Final Thoughts

AI generates almost 1000 lines of code. Of course, it has errors, but it would not need us if it did not, right?

If you want to follow AI news, use our assistants and catch the future, visit our platform, where you will have them all.

Here are the free resources.

Here is the ChatGPT cheat sheet.

Here is the Prompt Techniques cheat sheet.

Here is my NumPy cheat sheet.

Here is the source code of the “How to be a Billionaire” data project.

Here is the source code of the “Classification Task with 6 Different Algorithms using Python” data project.

Here is the source code of the “Decision Tree in Energy Efficiency Analysis” data project.

Here is the source code of the “DataDrivenInvestor 2022 Articles Analysis” data project.

“Machine learning is the last invention that humanity will ever need to make.” Nick Bostrom