Proper ML modeling starts with quality data.✅

Early on, you're going to have to tell between quality data and garbage.✅

Subject matter experts help make that distinction.✅
(Much before your feature engineering wizardry comes into play.)

A note about SMEs...

1/4

A dedicated data team to guide you is not guaranteed.

In my experience, SMEs were mostly non-technical.

Usually, the closest point of contact to the customers.

They tell you:
· "how" the data flows in the big picture.
· "why" some data is more significant than the rest.

2/4

Additionally, they help with:
· info on historic data.
· data labels if needed.

You gain a deeper understanding of data you'll work with.✅

Their domain knowledge goes beyond what you learn from Data models and Documentation.✅

3/4

So track down those SMEs and pick their brains throughout the model lifecycle!

Thanks for reading!

Please leave a like on first tweet and follow me @farazmunshi for more on ML and AI.

See you!

4/4