A scalable A/B testing backend using AWS Lambda and Redis HyperLogLog for cost-effective, privacy-conscious experimentation.
Gimel is an open-source A/B testing backend that enables developers to run and track experiments using AWS Lambda and Redis HyperLogLog. It provides a scalable, cost-effective solution for collecting and analyzing experiment data without storing personal information. The system handles event tracking and unique counting to measure variant performance in controlled tests.
Developers and data teams needing a self-hosted, privacy-compliant A/B testing backend for web or application experiments. It suits those familiar with AWS services and Redis who prefer control over their experimentation infrastructure.
Gimel stands out by combining serverless scalability with Redis HyperLogLog's memory efficiency, reducing costs and complexity. Its open-source nature and focus on aggregated, anonymized data offer a transparent, ethical alternative to proprietary A/B testing platforms.
Run your own A/B testing backend using AWS Lambda and Redis HyperLogLog
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
Uses AWS Lambda to automatically scale with traffic, handling varying loads without manual setup, as per the scalable architecture feature.
Minimizes expenses through serverless computing and Redis HyperLogLog's fixed 12k memory footprint per counter, reducing infrastructure costs.
Automates AWS Lambda and API Gateway setup with `gimel deploy`, simplifying deployment without complex configuration steps.
Aggregates anonymized data only, aiding GDPR compliance by avoiding personal information storage, as emphasized in the privacy section.
Provides a dashboard via `gimel dashboard` for visualizing experiment results, eliminating the need for external analytics tools.
Relies on Redis HyperLogLog which has inherent inaccuracies, especially for large experiments, as acknowledged in the README with a link to issues.
Tightly integrated with AWS Lambda and Redis, requiring users to manage and pay for these services, adding complexity for alternative setups.
Limited to fundamental A/B testing with event tracking; lacks advanced experimentation features like multivariate tests or detailed analytics found in commercial platforms.
Tracking endpoints may be blocked by ad-blockers due to URL keywords, necessitating custom endpoint configuration to avoid interference.