What project options do you have for a fellow newbie with little expertise in the DS field?
I want to have a thorough understanding of DS while working on this project.
2 Likes
DS is an overly wide phrase to suggest a project topic. What are your particular interests? Analytics, machine learning, artificial intelligence, or data modeling and engineering?
2 Likes
Machine Learning/ Artificial Intelligence
2 Likes
Nice…for ML you can start with unsupervised learning projects. E.g., predicting best combination of football players based on certain features (which involves k means clustering techniques) or something on those lines. In AI maybe you can start with generative AI using langchain framework which is easy to start and create a chatbot of your own 
2 Likes
Thanks, I will try to go with it.
1 Like
I would propose that you focus your job interviews on a certain industry or one in which you are actually interested (for example, marketing, finance, healthcare, etc.). From there, locate an open dataset relating to that industry and select a big business problem pertinent to the field, such as churn rate prediction or anomaly detection. If there are no free datasets available, consider using your own web scraper to collect data (for example, collecting product photos to forecast flaws).
Once you’ve gathered your data, try to create an end-to-end project that demonstrates a variety of your abilities. While beginning with notebooks to experiment is acceptable, I strongly advise gradually moving your codebase into more structured Python modules for things like data cleaning, investigation, display, and machine learning inference. To showcase your software engineering skills, employ object-oriented programming concepts. You should also think about creating an API as a backend and a Streamlit front-end to present everything in a clear, intuitive graphical user interface.
Make a GitHub repository and push your work gradually during the project. Using pipelines for continuous integration and deployment is also a smart option if you want your project to be automatically deployed to the cloud. If you’re feeling really ambitious, you can even further integrate unit tests into your continuous integration pipeline and use Docker to containerize your application.
The extent of your project is ultimately up to you. You will learn more and gain more skills to show prospective employers the more features you install. But even obtaining a workable project deployed to the cloud is already a noteworthy accomplishment that will make a good first impression on potential employers.
Wishing you luck!
1 Like
Thanks, I’ll mind and research about it.