Skip to main content
Link
Menu
Expand
(external link)
Document
Search
Copy
Copied
Home
Data Science Project Documentation Guidelines
Project Workflow Guide
Project Outline Template
Exploratory Data Analysis (EDA)
Getting Started With Exploratory Data Analysis (EDA)
Getting Started With Geographical Exploratory Data Analysis
Making Sense of the Census
How to Present Data Science Findings to Clients -- 3 Principles and a Case Study
Effective Data Visualizations
Topic Modeling
Web Scraping
Requesting Compute Resources
Spark! Guidelines
Github Administration
Git Conventions
Code Review Manual
Trello + Github Sync
How to Help
Spark! Project Organization
Slack
Spark! Tech Stack
Software Engineering Tech Stack
Web Frontend
Mobile Frontend
API/Servers
UI/UX Tech Stack
Data Science Tech Stack
Spark! FAQs Overview
Spark! FAQs
Tech Hiring Timelines
Anything Missing?: Submit a PR!
Spark! Tech Stack
Software Engineering Tech Stack
Universal Features
All projects should ideally be configured with the following solutions:
Each project should have a linter configured and applied at commit and or on push to github Automatic build and deploy(CI/CD) on push to production branches
Table of contents
Web Frontend
Mobile Frontend
API/Servers