OpenSource Projects

GreenGovRAG

GreenGovRAG is an AI-powered Retrieval-Augmented Generation (RAG) platform designed specifically for Australian environmental and planning regulations. It lets users ask natural-language questions and delivers accurate, citation-backed answers. The system supports geo-aware filtering (state, LGA, region), ESG-related queries (climate, water, biodiversity, emissions), and uses hybrid search combining BM25 and vector similarity. It is cloud-agnostic (AWS, Azure, Docker), supports multiple LLM providers, and is fully open source for contributors, civic-tech groups, and researchers.

Empowers planners, consultants, and researchers with accurate, location-aware answers to complex Australian environmental and planning regulations in seconds instead of days.

Stack: Python, FastAPI, REST, PostgreSQL, Qdrant, React, LLMs, LangChain

References - Application | dev.to | GitHub

Transtats

The project helps developers to ship RHEL/fedora packages with translation completeness, managers to automate localization workflow and oversee the localization effort progress by product releases as well as language maintainers to estimate translation volume.

Reduced translation delivery time by 20%, saving Red Hat localisation teams hundreds of hours annually and improving progress visibility for the language community.

Stack: Python, Django, REST, Celery, PostgreSQL, PatternFLY, JQuery

References - Fedora Magazine | Fedora Docs | GitHub