Digital Employee · Teacher
Arthur.
The teacher. Powers ArthurAI, the Agentic Learning Operating System — four editions in market today, with a fifth on the way.
Arthur is the Digital Employee for education. He is the runtime expression of the Eve-Education compound reasoning model — a domain-specialized Eve-Fusion architecture trained on the Eve-Genesis Education Edition synthetic reasoning corpus.
Arthur powers ArthurAI, MindHYVE's Agentic Learning Operating System. ArthurAI ships in four institutional editions today: the School Learning Edition (SLE) for K-12 districts and community schools, the University Learning Edition (ULE) for higher-education institutions, the Vocational Learning Edition (VLE) for vocational training authorities and workforce-development bodies, and the Corporate Learning Edition (CLE) for enterprise and corporate L&D. The Personal Learning Edition (PLE) for self-directed individual learners is on the way — same Eve-Education compound architecture that powers the four institutional editions, calibrated for the solitary user.
Arthur supports differentiated instruction, formative assessment, curriculum alignment, and the kind of structured pedagogical reasoning that historically only the most resourced school systems could afford. He inherits the universal Digital Employee properties — continuous learning, infinite memory recall, compliant posture, deployment on Eve-Grid on Microsoft Azure, global accessibility with FERPA and COPPA-aware behavior, and 24/7/365 availability.
Not just the lesson. The whole craft of teaching.
A teacher is never only a classroom instructor. Across K-12, university, vocational, and corporate learning, a teacher carries many roles at once — instructor, curriculum designer, compliance steward, institutional strategist. Arthur's agency spans all of them, in three orders of capability. Classroom differentiation is where it starts, not where it stops.
First order
Classroom instruction & differentiation
The teaching itself: differentiated instruction tuned to learner readiness and modality, formative assessment across knowledge, skill, and application, live scaffolding and remediation, source-grounded explanations and worked examples, and real-time progress tracking per learner.
Second order
Pedagogical design & institutional compliance
The work around the teaching: curriculum alignment to state and institutional standards, learning-pathway design, adaptive assessment calibration, FERPA- and COPPA-aware data stewardship, accessibility and universal design for learning, and the governance posture each jurisdiction requires.
Third order
Educational strategy & systemic scale
The strategy of the institution: learning-outcome modeling, curriculum coherence across programs, teacher professional development and competency tracking, and the access and equity goals an institution sets — the evidence base for scaling learning systems within and across institutions.
An AI with agency — and the architecture to wield it.
Arthur's agency is broad by design. What makes it deployable in schools, universities, and enterprises is the architecture beneath it and the Operating System above it.
Compound reasoning
Backed by Eve-Fusion™ F5
Every request runs through five models — a Phi-3 classifier, a Phi-4 reasoner trained on the Education Edition of Eve-Genesis, and three composed frontier models. Pedagogical reasoning depth no single model delivers.
Eve-Education F5 →Dedicated memory
Its own Azure AI Search
A dedicated Azure AI Search instance gives Arthur infinite memory recall across each learner's full record — every assignment, every formative result, every progress marker carried forward instead of forgotten between turns.
Bounded by the OS
Agency in the employee, control in the OS
Arthur reasons with full agency; ArthurAI — the Operating System — sets the bounds: FERPA and COPPA guardrails engineered into the reasoning substrate, a must-review-before-final gate on learner-data outputs, and a tamper-evident audit chain for every recommendation. The pedagogy is the teacher's; the data stewardship is the OS's.
One request, five models, one answer.
Eve-Education F5 is a compound, not a single model. Press Run to watch a request move through it — or click any stage to inspect it.
Stage 1 / 8 · Request
A user asks
A request enters through the Digital Employee’s Operating System — the surface that frames, scopes, and governs everything that follows.
The AI reasons. The human decides. The OS records.
“Agency in the employee, control in the OS” is the most important claim on this page — and the most abstract. Step through it.
Step 1 · Agency
Arthur reasons
Arthur works the case with full agency — reasons through it and drafts the learner-data report. No human in the loop yet.
One Digital Employee. Four institutional buyer types, plus the individual learner.
ArthurAI ships in four institutional editions today — SLE, ULE, VLE, CLE — each calibrated for the institution it serves: pedagogy, compliance posture, integration patterns, deployment scale. The Personal Learning Edition (PLE) for self-directed individual learners is on the way, with the same Eve-Education compound architecture underneath.
School Learning Edition
K-12 districts and community education
Differentiated instruction, formative assessment, and curriculum alignment for K-12 classrooms. FERPA and COPPA-aware. Live in pilot at the Federal Directorate of Education in Islamabad (IMC Girls F-11/3 + IMC Boys Pakistan Town).
University Learning Edition
Higher-education institutions
Adaptive learning, personalized pathways, and assessment for the university tier. Live at California Northstate University (pharmacy education), Daystar University (Kenya), the Open University of Kenya, GIKI, KUST, and across the IUCEA network.
Vocational Learning Edition
Vocational training authorities and workforce development
Live with the Pakistan Engineering Council across its 250,000-engineer CPD framework, COTHM hospitality (12,000+ students across 16 campuses), and the Development Authority of LaGrange (Georgia) for advanced-manufacturing workforce upskilling.
Corporate Learning Edition
Enterprise and corporate L&D
Workforce upskilling, compliance training, and leadership development for corporates. Aligned with WIOA, Georgia Work Ready, and the Technical College System of Georgia funding pathways for US enterprise deployments.
Personal Learning Edition
Self-directed individual learners
Personal learning, coming next. The same Eve-Education compound architecture that powers the four institutional editions, calibrated for the self-directed learner. Register interest at arthurgrid.ai/ple/.
Students train on ArthurAI in the classroom. They walk into the teaching hospital and work alongside ChironAI at the bedside.
Same compositional fabric. Same Eve-Genesis training methodology. The University Learning Edition (ULE) prepares health-sciences students for AI-augmented clinical practice; ChironAI CDS is the production system they encounter once they cross into clinical rotation. One reasoning architecture, two surfaces, one continuous pipeline — first realized in MindHYVE's strategic collaboration with California Northstate University and its teaching hospital under construction.