Santiago Profile picture
5 Apr, 12 tweets, 2 min read
Many people who want to start with machine learning face a big hurdle:

They think they can't do it at their current company.

But more often than not, this is not the case. This is a thread about things you can do to get past this.

↓ 1/12
First, don't worry if your company doesn't have a machine learning engineer position yet.

Look at this as a good opportunity!

Nobody has any expectations about the job yet, so you'll get to set the pace.

↓ 2/12
Focus on doing the work. The actual position, title, compensation, and other details will follow later.

Here is where you need to get creative, and these are two different strategies that I've seen working.

↓ 3/12
The first strategy is to look at your company's internal processes and find a way to introduce machine learning.

A few examples:

• Can you predict customer churn?
• Can you forecast their revenue?
• Can you improve a marketing campaign?

↓ 4/12
Look at every department. Talk to people.

What are they doing? What problems are they facing? What data do they currently have?

Everyone wants somebody else to make their life easier. They will happily open their doors to you.

↓ 5/12
If you can't make any progress, it doesn't matter. Nobody was expecting anything, so you learn something and move on.

But what happens if you can crack it and find a better solution?

Things will change really quick for you!

↓ 6/12
The second strategy is to find ways to solve part of your own job in a better way.

For example, if you are currently parsing a bunch of text, maybe NLP is helpful.

Or maybe you can use a clever algorithm to optimize a specific process of your application.

↓ 7/12
Where can you use the machine learning algorithm that you just learned to replace a clunky piece of the software you are currently building?

No, machine learning is not for everything, but you can certainly change the game if you find a good fit for it.

↓ 8/12
There's something similar about these two strategies:

You need to lead the charge, find a good use case, and develop a solution.

This is not easy, but if you pull it off, you'll be changing your future.

↓ 9/12
Yes, sometimes, this may be like trying to fit a square peg in a round hole.

If that's the case, you may want to consider finding a different job that gives you space to apply machine learning.

↓ 10/12
Remember that, sooner rather than later, most companies will start incorporating machine learning one way or the other.

They haven't because they need somebody like you to take the initiative and find a good use case.

↓ 11/12
Your goal is to have the space to keep learning and the ability to apply all of that in real-life applications.

Sometimes, finding a company that gives you this is what you need, but more often than not, you can make it happen right where you are.

Give it a try!


• • •

Missing some Tweet in this thread? You can try to force a refresh

Keep Current with Santiago

Santiago Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!


Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @svpino

6 Apr
There are a lot of moving pieces on a machine learning system.

This is a thread covering the backbone of the process, from data engineering all the way to a retraining pipeline.

Let's start. ↓
Everything starts with a problem you want to solve.

For example, you want to predict your company's sales in the next 12 months, and you have the last two years' worth of sales in a database.

When use case and data align, you are good to go!

The first step is to prepare the data to train a machine learning model that predicts future sales.

You have the data already, but you may need to transform it into a format that's easier for the model.

This process is called "Data Engineering."

Read 28 tweets
4 Apr
Learning a new language is not an obvious decision, especially when you are just starting in the industry.

Here are 10 frequently asked questions about learning Python 🐍. Hopefully, these give you the answers you are looking for.

1. Can I learn Python for free?

Yes. There are multiple YouTube videos, tutorials, and courses that will teach you Python for free.

But if you can afford it, I'd recommend you find a good MOOC that gives you some structure.

↓ 1/10
2. Is Python hard to learn?

It's not, especially compared with other languages out there.

That being said, becoming an expert is a life-long journey.

But one year of experience is more than enough for you to do whatever you decide to do.

↓ 2/10
Read 14 tweets
3 Apr
The Python 🐍 community on Twitter is amazing!

If you are a Python developer or you are looking to get started, introduce yourself below and let others connect with you👇
Hi 👋, I'm a machine learning engineer, and I've been coding exclusively with Python for 7 straight years.

I believe that Python is one of the most versatile languages you can learn today, and it's an investment with the potential to change your life.
The best part about this are the connections that this enables.

People saying hi, making study groups, asking questions, and helping each other.

Make sure to look through the comments. A lot of likeminded people willing to partner with you and do this together!
Read 4 tweets
3 Apr
25 True|False machine learning questions that are horrible for interviews but pretty fun to answer.

Most importantly: they will make you think and will keep your knowledge sharp.

These are mostly beginner-friendly.

1. A "categorical feature" is a feature that can only take a limited number of possible values.

2. Precision is a performance metric that defines a classification model's ability to identify only relevant samples.

3. Recall is a performance metric that defines a classification model's ability to identify all relevant samples.

4. One-hot encoding is an excellent solution to transform categorical features with high cardinality.

Read 14 tweets
2 Apr
You want to build a function to retrieve a value from a sequential list of unordered elements.

What would be the best approach?
You can assume that the size of the list is unknown.

Oh, sorry if this was confusing.

By "sequential list" I meant that elements come one after the other in memory. Think of a regular array.

It doesn't mean that you can't access elements out of order.

Read 4 tweets
2 Apr
When we start with machine learning, we learn to split our datasets in testing and training by taking a percentage of the data.

Unfortunately, this practice could lead to overestimating the performance of your model.

Imagine a dataset of pictures with people doing signals with their hands.

As we were told, we take 70% of the images for training and the remaining 30% for testing. We are careful to maintain the original ratio between classes.

How could this be a problem?

There are a lot of pictures of Mary in the dataset. She is showing different signals with her hands.

Also Joe. He was a model too that participated in the creation of the dataset.

Read 10 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!