Shipped
NRS expansion from rijāl works
New PIDs from Tahdhīb, Mīzān al-Iʿtidāl, and Jarḥ wa Taʿdīl grew the narrator reliability set to 26,923 indexed (21,617 reliability-graded) — lifting grading across the corpus. Delivered in v4.0.0.
بِسْمِ ٱللَّٰهِ ٱلرَّحْمَٰنِ ٱلرَّحِيمِ
The methodology is the product.
Live since March 2026 · Five locales · No signup required
We build tools that shorten the distance between question and evidence. Human judgment remains central. AI does the indexing.
EBO, never fatwa Theo produces Evidence-Based Opinions (EBOs) — structured analyses with evidence presented openly, reasoning visible, confidence calibrated, every source cited. Never a ruling. The scholar decides.
BY THE NUMBERS
Real platform. Real numbers.
Live counts from the production database, refreshed on every build. The home-page proof banner above and the cards below read the same stats endpoint; values match by design.
600,000+
Theo members worldwide
Cumulative Theo accounts since the platform went live in March 2026. Same count surfaced in the home-page proof banner, floored here to the nearest thousand.
1,990,000+
Citations served
Every primary-source reference Theo has anchored a claim in. Each hadith answer typically attaches the matn source plus the chain of narrators — roughly 18× the question count by design.
9.6M+
Pages of Islamic reasoning
Cumulative tokens Theo has generated, converted to paperback page equivalents at ~700 tokens per page. Anyone counting can do the math: pages ≈ tokens ÷ 700.
617,912
Hadith cataloged
Distinct hadith records cataloged by the IPSC pipeline across 123 verified primary-source works, including the Nine Books, Mu‘jam, Muṣannaf, Mustadrak, and fabrication catalogues. Of these, 480,334 (≈77.7%) are independently graded.
The first three cards are read at build time from the live stats endpoint (/api/v1/config/stats) — the same source the home-page proof banner reads. The fourth is a constant of IPSC v4.0.0. See the full changelog for cumulative release activity and the map below for geographic distribution.
GLOBAL REACH
Muslims learning worldwide
TheoAI is an open product — anyone can ask, no signup required, in every country. The plate below is rendered from real session data and refreshed live from the production API.
Sessions · last 30 days
—
Countries reached
—
Continents
—
Cohort live since
March 2026
Source: production sessions table, country code derived from request IP. Sessions include anonymous browsers and authenticated users (no signup is required to ask Theo). Live feed via api.theogrid.ai/v1/config/stats/geo.
WHO WE ARE
Built by MindHYVE.ai
TheoAI™ is MindHYVE.ai's Agentic AI for Islamic knowledge — purpose-built to serve Muslim communities with transparent, evidence-grounded tools. MindHYVE.ai — our parent company — was founded in 2022 by Bill Faruki and is now a 202-person agentic AI lab with operations across the United States, Pakistan, and Africa, partnerships in education (CNU, LaGrange, Open University of Kenya), and sister products in healthcare (ChironAI™) and legal services (JustineAI™). TheoAI is built on the same Eve-Grid™ agentic infrastructure that MindHYVE brings to those partnerships.
MindHYVE.ai
Purpose-built Agentic AI for regulated industries. A 202-person Agentic AI company building Agentic AI Operating Systems and the Digital Employees that run them — in healthcare, law, education and theology. Pioneering agentic AI since 2022.
Eve-Fusion™ cognitive engine · Eve-Grid™ infrastructure · Eve-Genesis™ training data
The MindHYVE portfolioVertical depth on shared cognitive infrastructure
Vertical depth on shared cognitive infrastructure.
One reasoning architecture. Thirteen editions generally available across five genuinely-specialised Agentic AI Operating Systems, running in five regulated verticals.
- Agentic Learning OS
ArthurAI™
For institutions that teach — K-12 districts, universities, vocational training authorities, corporate L&D.
Reasoning modesanalogical · Socratic · phenomenological
MottoVisit ArthurAI →The AI reasons. The educator decides.
- Agentic Healthcare OS
ChironAI™
For practices and health systems — reasoning-first clinical decision support and occupational medicine.
Reasoning modesabductive · analogical
MottoVisit ChironAI →The AI reasons. The clinician decides.
- Agentic Legal OS
JustineAI™
For personal-injury law practices and the institutional workflow they actually run.
Reasoning modesanalogical · abductive · dialectical
MottoVisit JustineAI →The AI reasons. The attorney decides.
- Agentic Finance OS
EliAI™
For desks that commit capital against evidence — underwriting, valuing and servicing illiquid claims.
Reasoning modesabductive · analogical
MottoVisit EliAI →The AI reasons. The underwriter decides.
Each product is a genuinely-specialised reasoner trained on its own Eve-Genesis edition. Same architecture. Different cognitive shape. One company that can credibly serve institutions across four regulated verticals — because the methodology that produces each reasoner is shared, and the application of it is not.
STRUCTURE
How we're organized
Parent Company
MindHYVE.ai, Inc.
Nevada C-Corporation. Product vision, AI research, and public-facing brand across all verticals.
Islamic Division
Eve-Theology, LLC
Dedicated subsidiary for Islamic scholarship products, corpus development, and community engagement.
Platform
TheoAI™
Three products — Theo (individual), Majlis (cohort), Mizan (enterprise) — all built on the same corpus.
MINDHYVE.AI PARENT-COMPANY PARTNERSHIPS
Real technology. Real institutions.
TheoAI is the Islamic-scholarship platform of MindHYVE.ai. Below is MindHYVE.ai’s verifiable track record across education, healthcare, and infrastructure — partnerships, launches, and offices on the same agentic-AI stack TheoAI is built on.
March 2026
California Northstate University
ArthurAI™ ULE + ChironAI™ pharmacogenomics CDS
Strategic Collaboration Framework to bring a reasoning-first agentic AI platform into the College of Pharmacy curriculum, with AI fluency certification through CIAI’s AI Professional 360™. Going live soon.
Press releaseApril 2026
Development Authority of LaGrange, Georgia
ArthurAI™ VLE + CLE workforce training
Live: adaptive workforce training across LaGrange’s manufacturing and industrial sector — Georgia International Business Park (GIBP) hosts 41+ international companies including Kia.
Press releaseJanuary 2026
Open University of Kenya
ArthurAI™ MOU
Strategic Collaboration MOU to integrate ArthurAI into OUK’s digital learning ecosystem.
Press releaseOctober 2025
California Institute of AI (CIAI)
AI Professional 360™
AI-authored curriculum: 18 courses, 1,000+ dynamically generated modules focused on AI literacy and fluency.
Press releaseQ4 2025
Africa operations — Nairobi
MindHYVE.ai regional expansion
MindHYVE.ai operates from offices in Newport Beach, Islamabad, and Nairobi. East African research and institutional partnerships are coordinated from Nairobi.
Press releaseSeptember 2025
ChironAI™ — Global launch
Agentic AI Clinical Decision Support
Intake, exam, labs, imaging, plan, documentation. Physician-attested on every clinical artifact.
Press releaseIN THE PRESS
Coverage.
- PRNewswireCNU and MindHYVE.ai Deploy First Agentic AI Learning Platform in U.S. Health Sciences Education2026-03-10
- GovTechAgentic AI Platform to Personalize Pharmacy Education at CNU2026-03
- PRNewswireMindHYVE.ai and the Development Authority of LaGrange Announce Strategic Collaboration on AI-Powered Workforce Development2026-04
- PRNewswire APACMindHYVE.ai Unveils ChironAI: World’s First Agentic AI Clinical Decision Support System2025-09
- PRNewswireMindHYVE.ai Brings Agentic Intelligence and Agentic AI Innovation to Africa’s Premier Tech Summit2025-05-26 — Connected Africa Summit, Diani, Kenya
- PRNewswireMindHYVE.ai and The Open University of Kenya Announce Strategic Collaboration on AI-Powered Learning2026-01
- TheCEOViewsBill Faruki: Pioneering Agent-Based Intelligence for Enterprise Transformation — Best CEOs to Watch 20252025
Every press item links to the source. List refreshed quarterly.
RESEARCH DIVISION
HYVE Labs
Eve-Fusion™
Cognitive reasoning engine powering multi-step theological analysis
Eve-Grid™
Azure-native infrastructure for secure, scalable AI deployment
Eve-Genesis™
250,000 training examples built from scholarly consensus
IPSC v4.0.0
667,153 documents across Quran, Hadith, Tafsir, and Fiqh
WHAT WE BELIEVE
Four commitments.
Transparency
Every answer surfaces its sources, confidence level, and limitations so scholars can judge independently.
Scholarly Rigor
Built for adab with texts — hierarchy of evidence, attribution, and careful language are non-negotiable.
Institutional Trust
Governance, data isolation, and audit-friendly architecture are first-class priorities, not afterthoughts.
Evidence-Based Reasoning
Reasoning is grounded in corpus operations. Human review remains central where stakes are high.
ROADMAP
What we’re shipping next.
v3.28
ʿIlal ingest (negative evidence)
al-Dāraquṭnī (4,348 entries) + Ibn Abī Ḥātim (638) + Tirmidhī al-Kabīr (1,264). Each ʿilal entry is a chain a classical scholar called out as defective — the negative-evidence layer.
v3.29–v3.32
12 more primary collections
Daylamī Firdaws, Ibn Saʿd Ṭabaqāt, Nasāʾī Sunan al-Kubrā, Bayhaqī al-Sunan al-Ṣaghīr, Bayhaqī Asmāʾ wa Ṣifāt, Ibn Wahb Jāmiʿ, Maʿmar Jāmiʿ, Ṣaḥīfat Hammām, Mullā ʿAlī Asrār, Abū Nuʿaym Maʿrifat al-Ṣaḥābah, Abū Nuʿaym Ḥilyat — each adds 5K–15K records.
Azure v4 cutover
Atomic alias swap to v4.0.0
Re-export tier files with the v4.0.0 corpus, refresh embeddings (588,737 re-embedded), atomic alias swap. Customers see no downtime; query URLs unchanged. Citation reproducibility preserved (prior backing indexes remain queryable directly).
v4.0+
Mega-targets & full re-cluster
Ibn Jarīr Tahdhīb al-Āthār (7,092 records), Ibn Abī al-Dunyā corpus (~50 small works), al-Khaṭīb Tarīkh Baghdād, Ibn ʿAsākir Tarīkh Dimashq (90 MB sharded multi-week). Plus full re-cluster of all 617K+ records (clusterAssignment 39.2% → 95%+).
Public roadmap. Updated as releases land. See thecorpus changelog for shipped releases.
SECURITY
Trust infrastructure.
Eve-Grid™ (Azure-native)
Microsoft Entra ID auth, Key Vault for all secrets, Private Endpoints for service-to-service traffic, App Insights observability
AES-256 Encryption
Data encrypted at rest (Azure-managed keys) and in transit (TLS 1.2 or later)
GDPR & CCPA
Cookie consent v2 banner; data export and erasure on request; consent-gated analytics
Patent Pending
US patent applications filed for matn-criticism pipeline, NRS synthesis, and classical-dimension detectors
REACH US
Get in touch.
MindHYVE.ai, Inc. · Newport Beach, California, USA
QUESTIONS
What investors and institutional buyers ask
Who builds TheoAI?
TheoAI is the Islamic-scholarship division of MindHYVE.ai, Inc., a 202-person agentic AI lab founded in 2022 by Bill Faruki with operations across the United States, Pakistan, and Africa. The corpus and pipeline are built by the Eve-Theology, LLC team. TheoAI is one of several MindHYVE products alongside ArthurAI™ (education), ChironAI™ (healthcare), and JustineAI™ (legal).
How do you grade hadith?
A computational implementation of the classical Ibn al-Salah five-condition framework: continuous chain (ittiṣāl), narrator integrity (ʿadālah), narrator precision (ḍabṭ), absence of irregularity (shudhūdh), absence of hidden defects (ʿillah). Each narrator carries a 12-tier reliability score (NRS). Disambiguated by the Persistent Identifier (PID) tiebreaker pipeline, structurally validated, scholar-in-the-loop pending on residue queues. See ipsc.theogrid.ai/methodology for the 1,436-line technical specification.
Has TheoAI been independently audited?
Yes. IPSC operates under a continuous quality-and-audit cycle: a 33-test regression suite, a 7-test cross-field consistency check, the NAQD-1/2/3 adversarial framework, and a recurring external-examiner cadence. The most recent cycle drove a v3.11 methodology recalibration. The full audit-and-quality practice — pillars, recalibration timeline, and the open scholar-collaboration program — is published at ipsc.theogrid.ai/honesty.
What is the relationship between TheoAI and MindHYVE.ai?
MindHYVE.ai, Inc. is the parent company; Eve-Theology, LLC is the wholly-owned subsidiary that operates the IPSC corpus and the TheoAI product. TheoAI is built on the same Eve-Grid™ Azure-native infrastructure, Eve-Fusion™ reasoning engine, and Eve-Genesis™ training methodology that power the rest of the MindHYVE portfolio.
Where can I see the roadmap?
The Roadmap section on this page lists v3.27 NRS expansion, Phase 4 ʿIlal ingestion, Phase 6 12-collection expansion, and the Azure v4 corpus cutover. The full version-by-version timeline lives at /changelog (product) and ipsc.theogrid.ai/changelog (corpus).
بِسْمِ ٱللَّٰهِ ٱلرَّحْمَٰنِ ٱلرَّحِيمِ
What's your question?
I'm here. Ask me anything about Islam. Every source cited. Every chain parsed.
Ask about any hadith, topic, or scholar…617,912 hadith cataloged · 480,334 graded (77.7%) · 95.8% Bukhari convergence