Data Scientist vs Data Analyst vs ML Engineer

For those of you who are still confused

I can give you two analogies to explain the difference between:

Data Scientist
Data Analyst
Machine Learning Engineer

Technical Analogy

In a digital product in the form of software, there are usually certain features that are directly related to customers. For example, there is a SaaS (Software as a services) that offers Trading Bot and Chart.

As a user, you can see trading data smoothly, in realtime — that is thanks to the hard work of a Data Engineer. They prepare the pipeline so data can move from point A to the user smoothly.

This SaaS provides various features for analyzing government policies, stock volatility, the latest issues, with various cool mathematical approaches. All of that is prepared by a Data Analyst. They provide visualizations that are easy to read, overviews that are easy for ordinary people to understand, and also provide all report related to variables that influence a stock price.

So what does a Data Scientist do???

Well, there is one more feature we haven't discussed yet, namely forecasting and stock price recommendations to users. All models built using machine learning methods up to cool deep learning are our job as a Data Scientist. Most data scientists will depend on Data Engineer and Data Analyst, for what?

Data scientists need analysis of a stock, they can ask a Data Analyst, or even do the analysis themselves. That is why many Data Scientist requirements demand statistical and mathematical skills
Data scientists need help with the model training flow and getting clean data. They can ask a Data Engineer to build a data flow pipeline to obtain clean data

If we look at the skill set, the comparison is roughly like this:

A Simpler Analogy

🟦 Data Engineer = Ingredient Buyer
They take all the raw ingredients (vegetables, meat, spices) from the market, then arrange them neatly in the kitchen. If the ingredients aren't there yet, the chef can't cook!

🟩 Data Analyst = Chef
They cook those ingredients into a delicious meal and arrange it nicely on the plate. They tell what has happened — for example, "This month, the best-selling item is meatballs!"

🟥 Data Scientist = Flavor Detective
They taste and predict the future — "Looks like tomorrow when it's raining, people will want to eat something warm!"

Remember the statistician in disguise from earlier? Well, that flavor detective is the disguised statistician!

The difference is easy to remember using question words:

Ingredient buyer → "Where are the ingredients?"
Chef → "What happened?"
Detective → "What will happen?"