Discover more from DSBoost
What do you need to get hired as a Data Scientist? - DSBoost #4
Welcome to the fourth issue of DSBoost, the weekly newsletter where you can discover interesting people in the ML/AI world, get the main takeaways of a relevant podcast, and stay up to date with the latest news in the field!
💬 Interview of the week
This week we interviewed Pauline, who is a Data Scientist at IBM. Enjoy:
What did you study?
I studied IT engineering.
What are your favorite resource sites and books?
I like Towards Data Science. The articles are very well written. I don’t use many other resources.
What got you into your current role (portfolio, certification, etc.)?
I started to work 8 years ago, there wasn’t much education about DS at that time. I had a few courses about Big Data. IBM recruited me as a consultant in DS and that’s where I learned Data skills.
What do you enjoy the most in your work?
I like to listen to customers’ needs and find a solution to bring value.
What tools do you use the most / favorite tools?
Excel / Pycharm / Jupyterlab to explore data.
Powerpoint to present results.
Do you use ChatGPT or other Al tools during your work? If so, how do they help you? Do they change your approach to problems?
Yes, when I need to code simple scripts and I think ChatGPT will code faster than I will.
Also, I use it to review my English (I’m French) and correct me if I make mistakes.
But there are limitations in a professional context because you can’t share what you are working on (data, issues, project information etc.).
What is your favorite topic within the field?
NLP, I’m fascinated by this field even if I don’t know much about it.
I also like dataviz because it really helps to understand data context better.
Which one of the recent AI/ML models will have the most significant impact on the industry in your opinion?
In my opinion, multimodal models are the future. Being able to translate text to images, then create graphs. The possibilities are endless.
What are you currently learning or improving (topics you are interested in nowadays)?
I’m currently interested in learning more about NextJS to present better-looking and interactive data visualization.
What is the biggest mistake you've made?
Not exploring data well. The more you look at your dataset, the more you can see atypia and ask the business teams questions about it. Which can lead to interesting discoveries.
What is your most significant achievement? (preferably DS related)
Working on a project from A to Z: I started with the client’s needs and I stayed until the project was live in production. It was a system to detect laundering money.
Your bio says “Indie hacker by night” what does it mean? How do you manage these projects besides your regular job? What skills do you have that are related to this hobby?
It’s not easy, but I block time to write blog posts, and tweets, engage with my network and work on strategy. All I can do from a mobile phone. At the end of the day or during the weekend I code.
Skills related are marketing, networking, and coding.
You recently launched TwittExplorer. Tell us more about it. What is it? What issues do you solve with it? Tell us more about the development.
Twittexplorer is a solution to export all your Twitter data into a google sheet: bookmarks, followers, tweets. You have a backup in case you lose your account, you can track your follower count and you can easily retrieve information and get value from your data.
I wrote an article here about the whole story behind it.
🎙️ Podcast of the week
What is Data Science? by Build a Career in Data Science.
What do you need at most companies to get hired as Data Scientist?
R or Python
Some Statistics & Machine Learning modeling
Go beyond just presenting the numbers to someone!
A Data Scientist can help a company make decisions based on data, which can be helpful in avoiding mistakes.
Data Scientists should not take data and get value from it. They should have goals and then use data to achieve these goals. Goal first, then data.
Diversity is important in Data Science: One person cannot do all the tasks. It is important to have a diverse Data Science team to ensure that all areas of the data science process are covered.
Thanks for reading DSBoost! Subscribe for free to receive new posts and support my work.
🧵 Featured threads
🤖 What happened this week?
OpenAI just made things a lot more interesting! They bought AI.com.
HuggingFace released LoRA scripts for efficient stable diffusion fine-tuning that optimizes language models by reducing trainable parameters. This technique can also improve denoising diffusion models for image synthesis tasks.
Last year Twitter launched their subscription service called “Twitter Blue”, which verifies the users’ accounts with a blue check. Meta has announced that they are also testing a new subscription service that will allow Facebook and Instagram users to pay for a verified account.
👥 Under the radar
Here are a few words from Derrick Mwiti personally:
I am Derrick Mwiti, a machine learning developer advocate with Neural Magic. I help the company reach developers by creating machine learning content that educates developers and shows them how to integrate our tools into their ML stack. I am the author of various books, including my latest book on writing for data scientists. I also teach data science and machine learning on online platforms such as Udemy. You can find me on Twitter and LinkedIn. Happy to connect.
Now you have the chance to buy Derrick’s book with a 20% discount!
Use this promo code at the checkout: dsboost