Internship goal

Leverage a generative model such as Chat GPT to create a chatbot to find information about Dataiku software, using the Dataiku reference documentation, the Dataiku Knowledge Base, the Dataiku Developer Guide and any other relevant documentation. Expose it in the help center.

Detailed description

At Dataiku we’ve been working on enhancing the onboarding user experience and as part of this initiative we’ve developed a new help center. Currently we can search through all the available resources using a Google search API which is exposed in the help center and the internship aims to provide a chatbot to enhance the user experience.

For example we could ask the chatbot this question: Can we import Excel files in Dataiku?

We would expect to receive an answer similar to this:

In Dataiku, you can certainly work with Excel files as datasets.

In Dataiku DSS:

Go to the Flow tab.Right-click on a folder and select "New Dataset."Choose "File" as the format.Select your Excel file.Configure settings like sheet and header.Review the inferred schema.Click "Create."

Now you can use the Excel dataset in Dataiku for analysis, preparation, and machine learning.

During this 6 months internship, you will:

• Get familiar with Dataiku and its help center.
• Integrate an existing NLP model using the provided API in the help center front end.
• Develop the user interface of a chatbot.
• Eventually suggest other possible enhancements, potentially incl. enhancements to the model itself.
• Celebrate and party because our beloved users would then be able to use the chatbot in the help center.

Stack

• TypeScript, Vue.js 3, CSS
• Python for model enhancements