DataRebel.io - Open Data Library
MY ROLES
TIMELINE
May 2021 - Oct 2021
TOOLS USED
React, Python, RAG, LLMs, Recommendation Engine, Django, AWS, Heroku, Data Visualization, Data Processing, API Integration, Jupyter Notebooks
DESCRIPTION
DataRebel.io is a comprehensive platform designed to make data more accessible by creating a central repository where users can find existing analytics, insights, and data sources. The vision was to reduce redundant data work by enabling researchers, analysts, and businesses to discover, share, and build upon existing data projects rather than starting from scratch.
OUTCOMES
- Created a searchable repository of open datasets across various domains
- Developed standardized metadata schema for describing data projects and their insights
- Built easy-to-build user portfolios to encourage data sharing and collaboration.
FEATURES
Advanced dataset search, project showcasing, interactive data previews, job board, contribution workflow for community additions
CHALLENGES
Standardizing diverse datasets with varying formats and quality. Creating effective metadata to make data discoverable. Again, distribution and jumpstarting the network always ends up being the most challenging aspect. The product is not useful unless there is a lot of data and projects but without data and projects people won't use/have incentive to add their own projects/data.
APPROACH
Build out the basic platform to allow discovery of data, projects, and jobs. Then built out tools and features that incentivized people to share. ie. personalizable portfolios (like LinkTree/Github) and job recommendation enginge and custom cover letter generator based on the projects and insights you have shared.