Connect
Sign up to receive our monthly newsletter.
Inquiries
Have a question? Contact us here.
Job Opportunities
Learn about job openings.
Digital Public Goods Alliance - 2026
Sign up to receive our monthly newsletter.
Have a question? Contact us here.
Learn about job openings.
DPGs for AI Collection
All DPGs included in the DPGs for AI Collection must meet the requirements outlined in the DPG4AI collection criteria.
The information below reflects the self-reported responses provided as part of the assessment process.
Solution Type: Open Source Software
DPG Compliance & Profile Page: https://www.digitalpublicgoods.net/r/feluda
Description: Feluda is a configurable engine for analysing multi-lingual and multi-modal content. While flexible, it is built it to analyse data collected from social media - images,text and video. It is configurable to allow for different use cases such as finding trends and narratives in social media across languages, and surfacing fact checked content corresponding to image and video.
Assessment Status: Under Review
Review Date: 2026-06-14
Category Fit:
Open Source Software (GPL-3.0; Python). Feluda is a configurable engine for multilingual, multimodal content analysis (text/image/video/audio) built around swappable 'operators' (plugins). Lifecycle roles: Data Acquisition & Preparation (embeddings, hashing, text/entity extraction and indexing of social-media content) and Deployment & Integration (engine/microservices powering search, clustering and analysis). Verified DPG since 2023. Repo: https://github.com/tattle-made/feluda
AI Lifecycle Utility:
Documented Relevance/ Impact:
Powers Tattle's misinformation-analysis tooling — e.g. Khoj reverse-image search for fact-checks, and analysis of COVID-19 WhatsApp messages in India. Supports ResNet / Sentence-Transformer embeddings and pHash search for cross-lingual trend and narrative detection. Operators published on PyPI under the 'tattle' org. Docs: http://feluda.tattle.co.in/
Adoption Readiness Level:
L4 Orchestrated / Optimized
Adoption Readiness Evidence:
pip-installable core plus separately versioned operator packages (PyPI), YAML-configured engine, CI/CD, microservices architecture, contributor wiki. Active maintenance; DPG-verified since 2023. https://pypi.org/user/tattle/
Interoperability Level:
L4 Integrated — versioned API/SDK, plug-in capable, automated synchronization
Interoperability Evidence:
Plugin/operator architecture lets users swap implementations without lock-in (e.g. OpenCV instead of Google Cloud Vision; hashing instead of heavy ML). REST + JSON APIs; standard embedding models (ResNet, Sentence Transformers) and pHash. Operators are independently installable packages: https://github.com/tattle-made/feluda/tree/main/operators
Responsible Practices Level:
L2 Partial — some limitations mentioned without systematic coverage
Equity & Inclusion:
Responsible Practices:
Inclusion & Autonomy:
Responsible Practices Evidence:
README documents operator trade-offs (heavy ML model vs lightweight hashing; OpenCV vs proprietary Cloud Vision) which surface accuracy/cost limitations, but there is no consolidated limitations / bias / failure-modes statement. Recommend a dedicated LIMITATIONS section to reach L3.
