Challenges

Overview

Applicants to the Launchpad UpSkill program are required to work on four challenge problems. Each challenge has a distinct focus and will test your problem solving and organization skills. Carefully follow the instructions and pay close attention to the solution submission guidelines; poorly formatted solutions will count against you during the evaluation process.

Challenges

1-Spreadsheet Analysis

Analyze some data, answer some questions, and present your findings. Refer to the spreadsheet challenge specifics for more details.

2-No-code Machine Learning (Jeans)

Build a dataset and train a model using a no-code platform; use your model to categorize some jean pictures. Refer to the platform challenge specifics for more details.

3-No-code Machine Learning (Jackets and Shirts)

Build a dataset and train a model using a no-code platform; use your model to categorize images of jackets and shirts. Refer to the platform challenge specifics for more details.

4-No-code Natural Language Processing (Spam or Ham)

Label a dataset and train a model using a ready jupyter notebook; use your notebook and training data to make predictions. Refer to the Spam or Ham challenge specifics for more details.

Helpful Hints

Although arriving at a functional solution is important, we're more interested in how and why you arrived at your solution. You should be able to summarize and present your work while justifying and explaining the decisions you made along the way. Here are some things you may want to consider while working on your challenges:

  • Always look at your data.
  • Challenges are not trivial; they should take anywhere from a few hours up to a day to complete.
  • Read the instructions and the README.md files, and make sure you have a good understanding of the challenge problem, the data, and the desired solution before you start your work.
  • What types of visualizations will help me grasp the nature of the problem?
  • What types of visualizations and summaries will help me present my solution to a key stakeholder?
  • What are some of the inherent limitations to my solution, and how can the solution be improved to over come them?
  • If there are any, what are some of the business implications relevant to your solution?
  • Please note that the (2-No-code Machine Learning (Jeans)) challenge and (3-No-code Machine Learning (Jackets and Shirts)) challenge are very similar. You will use the same tools and strategies on two different data sets.

Deadline

Log in to your account and upload the public file sharing link via the dashboard. Don't put your file in a folder before compressing; compress the file itself. Challenges should be submitted no later than midnight PST on January 5 for all employees.