Quickstart Guide

Welcome to the mimilabs workspace! This guide will help you get started with the workspace.

Step 1: Sign Up

Once you sign up, you will receive two invites: 1) the Databricks account (via AWS Identity Provider) and 2) the mimilabs Slack group.

The minilabs workspace uses the Databricks data lakehouse. All the data and codes will sit on the Databricks view. Separately, we use Slack to interact with the user community. There, you can introduce yourself, share your interests, and get some help from like-minded colleagues. Don't be shy! We are gathering here to make sustainable changes to the US healthcare system.

After you receive a Databricks invitation (sent by AWS), here are the steps:

  1. Change your password at the first login.
  2. Set up your MFA device.
  3. Sign in to the workspace.

Regarding the Slack group, you will receive an invitation separately in your email.

Step 2: Get Familiar with Databricks

If you are already familiar with Databricks, please skip this step.

Also, don't worry if you haven't used Databricks before. The platform is straightforward to use, and many videos explain the company and its platform. For now, you don't need to know every function/module of Databricks. You need to get familiar with the data lakehouse concept and how we can navigate the vast datasets.

Step 3: Be friends with Databricks Assistant and mimi-bot

Large Language Models (LLMs) will forever change how we interact with complex data. They can help us find relevant datasets from the oceans of data, help us write queries, help us combine them, and help us come up with new and important research questions that we never thought of.

Through the guide, you will heavily rely on two LLM agents:

  • Databricks Assistant on Databricks - powered by OpenAI
    • It can help you find the right tables and write queries
    • Read this to know more about Databricks Assistant
  • mimi-bot on Slack - powered by Anthropic
    • It can also help you find the right tables and come up with interesting research questions
    • You can use mimi-bot to track usage volume and some other administrative tasks

Here is a quick demo on how you can leverage mimi-bot on Slack:

These two bots are similar yet different. We recommend you use both to get the most out of your work.

We suggest you utilize the power of LLMs as much as possible in your work even after you complete the tutorial.

Step 4: Watch mimilabs Tutorial Videos

We have created a series of tutorial videos to help you get started with the workspace.

The first video from the series is below:

Watch more videos by clicking the links below:

If you need any additional data sources or have any questions, please contact us at info@mimilabs.ai or ask your questions in the Slack group!

Resources

Data Catalog

Learn about the list of datasets loaded into our data lakehouse and how to access them.

Data Engineering

Learn about how we downloaded and ingested thousands of public datasets into our data lakehouse.

Pricing

Learn about the pricing model and how to estimate the cost of using our data lakehouse.

FAQs

Learn about the frequently asked questions and answers about our data lakehouse.