Transparent Data Sources — Where Our Content Comes From

Content at a Glance

The following figures reflect live counts from our production database as of July 2026. Each links to its source breakdown below.

Current Affairs Sources

Our current affairs are ingested from nine established sources — newspapers, government press releases, legislative research, think tanks, and legal databases. A Python-based ingestion pipeline, triggered daily via GitHub Actions (approximately 6 AM IST), fetches RSS feeds and article pages, then processes each article through a three-stage LLM pipeline:

1. 1Classification — Categorize each article by UPSC subject (Polity, Economy, Environment, etc.) and sub-topic.
2. 2Enrichment — Extract key facts, link to related schemes, PYQs, and geography features.
3. 3Scoring — Rank articles by UPSC relevance to surface the most exam-critical content first.

Over 10,000 articles have been indexed to date, each traceable to its original source URL and publication date.

Source: Vaidra current affairs ingestion pipeline (Python + GitHub Actions + LLM). Article count reflects production database as of July 2026.

Previous Year Questions (PYQs)

Every question in our PYQ bank — 21,061 questions — is sourced from official examination papers published by the conducting body. We do not generate, paraphrase, or fabricate questions. Each question is traceable to its exam name, year, and paper.

Source: Official question papers and answer keys from UPSC and State PSCs. Questions are extracted via automated parsers and verified against the original PDF. Total: 21,061 questions across 7 exam families.

Geography Lab

Our interactive geography lab contains 1,379 features across 10 layers, rendered on Mapbox GL maps. Every feature is provenance-backed — it cites a specific, verifiable source with a URL and a last-checked date. Users can see the source attribution for any feature directly in the map interface.

We do not generate geographic data from AI without grounding it in an authoritative source. When AI is used to enrich a feature's description, the underlying spatial data and classification always trace back to an official dataset.

Source: Official government and research datasets (CWC, WII, NCERT, ICAR, Census of India, Köppen-Geiger V3, IMD). Each feature carries a source URL and last-verified date. 1,379 features across 10 layers as of July 2026.

Learning Topics

Our learning module contains 1,831 topics drawn from the learning_topics table, each mapped to the official UPSC Civil Services syllabus. Topics cover all four GS papers (GS1–GS4), the Prelims syllabus, CSAT, and optional subjects. Each topic links to related current affairs articles, PYQs, and geography features — creating an integrated study network rather than isolated content silos.

Topics are structured hierarchically: broad subject areas (Polity, Economy, Geography, etc.) drill down into specific sub-topics that mirror the UPSC syllabus structure. This ensures that every piece of content on the platform is connected to the exam framework an aspirant is studying for.

Frequently Asked Questions

Where do the current affairs articles on Vaidra come from?

Our current affairs are sourced from nine established publications and government/institutional bodies: The Hindu, Indian Express, Press Information Bureau (PIB), PRS India, Ministry of External Affairs (MEA), Observer Research Foundation (ORF), Institute for Defence Studies and Analyses (IDSA), LiveLaw, and PIB Cabinet releases. A Python-based ingestion pipeline (triggered via GitHub Actions on a daily schedule) fetches RSS feeds and article pages, then applies a three-stage LLM processing pipeline — classification, enrichment, and scoring — to filter, tag, and rank each article for UPSC relevance. Over 10,000 articles have been indexed to date, each traceable to its original source URL.

Are the previous year questions (PYQs) from official exam papers?

Yes. Every one of the 21,061 previous year questions in our database is sourced from official examination papers. These span seven exam families conducted by the Union Public Service Commission (UPSC): Civil Services Examination (CSE), National Defence Academy (NDA), Combined Defence Services (CDS), Central Armed Police Forces (CAPF), Engineering Services Examination (ESE), Geo-Scientist Examination, and seven State Public Service Commissions (MPPSC, UPPSC, UKPSC, RPSC, HPSC, PPSC, and BPSC). We do not generate or fabricate questions — every question is extracted from the official question paper or answer key published by the respective examining body.

How is the geography lab data verified?

Our geography lab contains 1,379 features across 10 layers (rivers, water bodies, ecology, islands, mountains, industry, climate, agriculture, demography, and soil). Every feature is provenance-backed — meaning it cites a specific, verifiable source such as the Central Water Commission, Wildlife Institute of India, Census of India, NCERT textbooks, ICAR soil classification, or the Köppen-Geiger climate dataset. Sources are recorded with a URL and a last-checked date. Features are rendered on interactive Mapbox maps with source attribution visible to the user. We do not generate geographic data from AI without grounding it in an authoritative source.

How often is content updated?

Current affairs are ingested daily via an automated GitHub Actions pipeline that runs on a schedule (approximately 6 AM IST). The pipeline fetches new articles, processes them through the three-stage LLM pipeline, and writes them to our database. Previous year questions are updated whenever a new official exam paper is released and processed. Geography lab data is reviewed and updated as official sources publish new data (e.g., new Ramsar sites, new tiger reserves, Census updates). Learning topics are updated as the UPSC syllabus evolves. The static content on this page is revalidated every 7 days (604,800 seconds).

Do you fabricate testimonials or inflate user counts?

No. We do not fabricate testimonials, user reviews, or success stories. We do not display inflated subscriber counts, fake ratings, or manufactured social proof. If a number appears on our platform — whether it is article counts, PYQ counts, or geography feature counts — it reflects a real query against our production database. Our testimonials page only contains verifiable submissions. This transparency page exists precisely so that users, journalists, and researchers can verify our data claims independently.

Can I verify these numbers myself?

The counts cited on this page (10,000+ current affairs articles, 21,061 PYQs, 1,379 geography features, 1,831 learning topics) are derived from our production PostgreSQL database. While the database itself is not publicly queryable, every individual piece of content on the platform links back to its original source. Current affairs articles display their source URL and publication date. PYQs display the exam name, year, and paper. Geography features display their source citation and last-verified date. You can cross-reference any of these against the primary source. If you find a discrepancy, contact us at [email protected] and we will investigate and correct it.