Top five habits and tools to keep your data science technical skills sharp
Some of the links in this article are affiliate links. This means if you click on the link and purchase or subscribe, I will receive a commission. All opinions and recommendations remain objective. You can also read the affiliates disclosure for more information.
Recently I have been reviewing my process for continuous learning and practice to keep my data science skills sharp. To stay ahead, it's essential to adopt habits and leverage tools that foster growth in both technical and analytical skills. It's also crucial to take the time to revisit core topics and concepts you might not have used in a while.
For each habit I've provided a companion platform and explained why these platforms like DataCamp, StrataScratch, Kaggle, O’Reilly, and DigitalOcean can help you both revisit the basics and sharpen up those skills.
Keeping the basics sharp also helps with interview preparation - if that's something you're interested in check out Preparing for a statistical data science interview and Exploring coding interview topics in Python. I think staying technically sharp makes you more confident, job-ready, interview-ready and all round a better Data Scientist.
All of the habits and tools mentioned in this article have proven immensely useful to me. Let's dive in!
Learn with DataCamp
Why DataCamp?
DataCamp offers interactive courses tailored for data scientists at all levels, covering topics from data cleaning to advanced machine learning. Its bite-sized lessons and hands-on coding exercises ensure you learn concepts effectively and can apply them immediately. It also has a great mobile app which I've been using to practice in the evening during downtime. As an intermediate Data Scientist I've been using my DataCamp subscription to revisit basic topics like Supervised Learning with scikit-learn and advanced topics like Retrieval Augmented Generation (RAG) with LangChain and MLOps. The pricing is clear, there are fairly regular sales and discounts so watch out for those if interested, and there is a discount for a yearly subscription.
Other courses I have taken recently include:
- Machine Learning with PySpark
- Introduction to Deep Learning with PyTorch
- Retrieval Augmented Generation (RAG) with LangChain
Key Benefits:
- Wide range of courses covering Python, R, SQL, and more
- Skill tracks and certifications for guided learning paths
- Built-in coding environments to practice directly in the browser
Habit:
Dedicate 15–30 minutes daily to progress through a course or learn a new skill. This incremental approach builds consistency over time.
Full review:
For a more in-depth review on DataCamp and why it's great for continuous learning check out Developing your data science and analytical coding skills - a review of DataCamp.
Code with StrataScratch
Why StrataScratch?
StrataScratch is designed to help data scientists improve their coding and problem-solving skills with real-world SQL and Python challenges. The platform features interview-style questions sourced from companies like Google, Airbnb, and Facebook.
I really love this platform and took the option to purchase a Lifetime membership - great that they offer this in the age of subscriptions! Sometimes they have sales like 30% discounts so if you're interested watch out for those. Couldn't recommend it highly enough. Their frontpage states they have 1,000+ interview questions, 200+ companies tracked, the first 50 questions are free and new interview questions are released every month. What's not to love for staying sharp?
Key Benefits:
- Coding questions - Analytical, Algorithm and Visualisation questions with optional hints and a guided Solution section if you're struggling, Solutions from Users, Discussion area and Resources for learning
- Non-coding questions - Business Case, Modelling, Probability, Product, Statistics, System Design, Technical
- Data projects - Business Analysis, Classification, EDA, NLP, Regression, Clustering, Data Engineering
- Guides - SQL and Python data manipulation plus time and date manipulation
- Covers many language options - PostgreSQL, MySQL, MS SQL Server, Oracle, Python-Pandas, Python-Polars, pySpark and also R
- Database-focused questions that mimic real-world scenarios
- Tutorials and solutions for a better learning experience
- Excellent preparation for technical interviews
Habit:
Solve 2–3 coding challenges daily. Focus on SQL queries and data manipulation techniques, as these are core to many data science roles. Expand into Python data manipulation alongside data structures and algorithms questions.
Practice with Kaggle
Why Kaggle?
Kaggle is a cornerstone platform for data science practice. It offers everything from datasets and competitions to an active community where you can collaborate and learn from others.
Key Benefits:
- Participate in competitions to solve real-world problems
- Explore extensive datasets for hands-on practice
- Learn from notebooks and solutions shared by other users
Habit:
Start with smaller competitions or use Kaggle’s practice problems to familiarise yourself with its tools. Build a project portfolio by publishing your own notebooks.
Read with O’Reilly
Why O’Reilly?
O’Reilly’s vast library of books, videos, and live training sessions is an indispensable resource for any data scientist. Covering cutting-edge technologies, methodologies, and trends, it’s your go-to for staying informed. At the time of writing O’Reilly look to offer a free-trial to try out the service.
Key Benefits:
- Access to thousands of books and videos on technical topics, some of my favourites include:
- Regular live events and webinars led by industry experts
- Coverage of tools like TensorFlow, PyTorch, and more
Habit:
Allocate time weekly to read a chapter or watch a tutorial on a new concept.
Deploy with DigitalOcean
Why DigitalOcean?
Deployment is a critical skill for data scientists. DigitalOcean provides a straightforward and affordable way to host your models, dashboards, or applications in a production environment. Simple, predictible pricing with monthly caps and flat pricing alongside good documentation make DigitalOcean one for you to strongly consider when it comes to deploying solutions to the cloud. At the time of writing, there is an offer to try DigitalOcean free with a $200 credit. A great way to begin exploring cloud infrastructure!
Key Benefits:
- Simple interface for setting up virtual machines and cloud resources
- Pre-configured templates for deploying apps, including ML models
- Scalable solutions for professional-grade projects
Habit:
Use DigitalOcean to host a personal project, such as a dashboard or API, to understand the end-to-end workflow of deploying data science solutions. I will be releasing an article covering how to quickly deploy projects to DigitalOcean in the near future.
Enjoy the learning journey
By leveraging these platforms, DataCamp for learning, StrataScratch for coding, Kaggle for practice, O’Reilly for reading, and DigitalOcean for deploying, you can keep your data science skills sharp and versatile. Consistency is key and integrating these tools into your routine, and you’ll be well-equipped to tackle challenges and keep skills sharp in the field.
One final bonus resource I will mention is Ace the Data Science Interview: 201 Real Interview Questions Asked By FAANG, Tech Startups, & Wall Street which I've found to be an effective complete refresher across data science as a whole albeit focused on technical interviewing. On that note if you were preparing for a technical interview where the annoying data structures and algorithms (DSA) style questions pop up, probably should mention Leetcode to prepare for that - I don't find these DSA questions as practical as the main StrataScratch questions though!
I hope you enjoyed the article and as always be sure to check out other articles on the site. You may be interested in:
