Citizen Data Science campaign
summary
Three years of work integrate Jupyter into a highly regulated US intelligence environment, producing an open-source env gallery notebook-sharing platform and micro data science VMs that spin up personalized ephemeral Jupyter environments and forbid sharing outputs to preserve need‑to‑know security. The system records per‑cell execution to compute a health score that degrades after 30 days, and a recommender stack drives discovery where 15% of views come from recommendations; the deployment runs thousands of notebooks by hundreds of authors with a 6:1 user-to-author ratio.
AI-written summary · about
chapters (27)
- 0:00Introduction
- 0:16Agenda
- 0:35TLDR
- 1:04Intelligence Community
- 3:09Why Jupiter
- 4:31Challenges
- 4:53Data sensitivity challenges
- 6:06Micro data science environment
- 6:55Envy Gallery
- 8:02MB Gallery
- 8:15Notebook Gallery
- 9:05Recommenders in health
- 9:50Recommender analytics
- 11:15Enemy Gallery
- 12:14How effective is this
- 12:53Notebook health
- 13:47Why is notebook health important
- 16:14Notebook health graph
- 18:54Notebook adoption
- 20:11Notebook metrics
- 21:12Analyst advocates
- 21:55Use cases
- 22:05Operational notebooks
- 22:25Enterprise notebooks
- 23:12Building blocks
- 24:18Notebook breakdown
- 25:04Future efforts
related talks
01h 28mHacking the Data Science Challenge
34mKeynote: Lighthearted Thoughts on Conversing About Data
30mEmpowered by Python - A success story
25mScope of data journalism for developers
12mCommunity matters: 8 reasons why you should be involved with
29mFrom rocks to a hammer when and how to change your company's analytical tools
42mDevelop Data Science for Business Value: a Phoenix Story
15mFast-tracking your AI Career with Kaggle