The Quality of Projects in your Portfolio Matters a Lot when Applying for Data Scientist or Machine Learning Engineer Positions

Author(s): Suhas Maddali

Originally published on Towards AI.

The real-world impact that a project makes with the use of machine learning and data science can be a game changer when it comes to interviewing for data science and machine learning related positions.

There are many websites and blogs that highlight the importance of building portfolios to land your first machine learning or data science job provided that you do not have enough experience. While this could not be further from the truth, it is also important to consider how much of an impact these projects on portfolios have made in the real world if applied and used. There are many online courses that teach the fundamentals of machine learning and expect that potential candidate becomes an expert or proficient by teaching them how to perform exploratory data analysis, data processing, data preparation, and model predictions on toy datasets, with one of them being the famous Titanic dataset. I would say that it can be a good start and the online courses teach the right fundamentals needed for a student. But if they just assume that the real-world data can also be quite similar and no further action should be taken, this could be a false assumption.

A student can get to know the overall process of machine learning and how the data preparation is done, and many other aspects, which is a good thing. However, the real-world data can be quite complex with a large number of missing and messed up data values. Therefore, a significant amount of time should be spent on feature engineering alone to find the relevant information for the predictions.

Though it can be a good start with solving the toy datasets in the beginning, the actual data science acumen and skills are tested when solving some interesting real-world company challenges on various websites such as Kaggle. Solving some of these challenges can be extremely valuable so that candidates can get to know how the data processing is carried out and further can take the right steps to make the data much more acceptable for machine learning models. In addition to this, they can also discuss with others along the way and find out the best data processing mechanisms that ensure good performance by the models depending on the evaluation metrics.

If you are looking for some interesting projects that you might add to your machine learning and data science portfolio, feel free to take a look at my earlier article, where I highlight some of the interesting projects that can add quite a lot of value when you are building a strong data science portfolio. There can be other projects as well that you might add that you think can make an impact. There is a lot of data available on Kaggle, and companies are constantly looking for people who would be able to solve their problems and challenges with machine learning and data science. Below is the link to my earlier article.

Best Projects to Showcase in Machine Learning Portfolio U+007C by Suhas Maddali U+007C MLearning.ai U+007C Medium

Now that you have determined that build a portfolio with projects that solve some of the real-world challenges, the next part would be to determine which platform can be the best way to showcase the projects. There are many ways to showcase your portfolio with all of your work. One of the easiest ways to showcase that work is on GitHub. One suggestion would be to add a readme file at the very beginning of your portfolio page where you might post your additional content about the project and information about it. You can also feel free to take a look at my GitHub portfolio where I highlight a list of all the projects.

suhasmaddali (Suhas Maddali ) (github.com)

Lacking experience in the field of data science can sometimes have an impact, especially if applying for those roles that involve real-time predictions and deployment of machine learning applications. In such cases, it can be a good thing to actually build portfolios of your work and showcase those during the interview so that employers can also get to know how you have approached the problem and how relevant is it to the project that they have considered. If you are looking for ideas about where to start building the portfolio or looking for additional help, feel free to take a look at the video below, where I highlight how I have actually built my portfolio. You can also reach out to me on LinkedIn or Facebook so that I can help you in designing and implementing a solid portfolio.

Conclusion

There are a large number of courses and certifications that do a good job in highlighting the fundamentals of machine learning and data science. In order to be a great data scientist, however, it is always important to also solve some of the real-world challenges by taking a look at the problems that companies are facing with their data and how they are implementing machine learning and data science to solve their problems. After reading this article, you must have gotten a good idea about how to build a strong portfolio and have understood the importance of building projects that make real-time impact. Thank you for taking the time to read this article.

If you are interested to stay updated with the latest articles on data science, feel free to subscribe to my mailing list and also becoming a member on Medium where you can have access to unlimited articles without restriction.

Your membership fee directly supports Suhas Maddali and other writers you read. You’ll also get full access to every story on Medium. Click the link below to become a member on Medium with just 5 dollars per month and get access to unlimited list of articles. Below is the link. Thanks.

https://suhas-maddali007.medium.com/membership

Below are the ways where you could contact me or take a look at my work.

GitHub: suhasmaddali (Suhas Maddali ) (github.com)

LinkedIn: (1) Suhas Maddali, Northeastern University, Data Science U+007C LinkedIn

Medium: Suhas Maddali — Medium

Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.

Published via Towards AI