{"repo":"Mohith-akash/Global-News-Intel-Platform","free":true,"listed":false,"github":"https://github.com/Mohith-akash/Global-News-Intel-Platform","clone":"git clone https://github.com/Mohith-akash/Global-News-Intel-Platform.git","description":"AI-powered geopolitical news intelligence platform. Ingests 100K+ daily events from GDELT, stores in MotherDuck (DuckDB), orchestrates with Dagster, and features an AI chat interface with Text-to-SQL. Full data engineering stack at $0/month.","language":"Python","stars":23,"topics":["ai","cerebras","ci-cd","dagster","dashboard","data-analytics","data-engineering","data-pipeline","etl","gdelt"],"license":"MIT","category":"data-pipelines","readme_excerpt":"Global News Intelligence Platform Serverless ELT pipeline that ingests, processes, and visualizes 100,000+ global news events per day from the GDELT Project, with an AI chat interface for natural-language queries. Live dashboard: https://global-news-intel-platform.streamlit.app/ Metric Value -------- ------- Cumulative events processed 20M+ Daily ingestion 100K+ events Live operation 8+ months, continuous scheduled runs Unique visitors 6,000+ in the first few months, 100+ new daily, no promotion Coverage 200+ countries, 100+ languages Typical query latency < 1 second Monthly infrastructure cost $0 GDELT monitors news media from nearly every country in 100+ languages, identifying the people, locations, themes, and emotions driving global society. Architecture Data flow 1. Extract: GDELT Events API + GKG feed, parsed with Polars 2. Validate: schema and threshold checks before anything is written 3. Load: deduplicated inserts into MotherDuck (serverless DuckDB) 4. Transform: dbt models build staging views and mart tables 5. Emotions: GKG tone/fear/joy/topics extracted on a rolling 24h window 6. Embed: Voyage AI generates 1024-dim vectors every 12 hours 7. Serve: Streamlit dashboard with dual-mode AI chat (SQL + RAG) Cost engineering The pipeline started on a Snowflake trial. When the trial ended, the warehouse moved to MotherDuck and the slowest processing stage was rewritten from Pandas to Polars ( 10x faster), bringing the total monthly cost to $0 on free tiers, without giving","default_branch":null,"files":null,"tree":[],"storefront":"/r/Mohith-akash","claimed":false,"request_supported":{"post":"https://gitbuyer.com/r/Mohith-akash/Global-News-Intel-Platform/request-supported","requests":0},"note":"indexed from public GitHub; nothing is for sale on this page. Clone it from GitHub. Paid listings live at /search."}