Methodology AI Method Research For Teachers Projects Writing About Available for work Contact Glossary Free Resources Book a call — first session free
AI-Assisted Method
Students & teachers Systems engineers & ed-tech developers

Method-first AI.
Not a chatbot that knows math.

Most AI tutoring tools hand students answers. What I built is different — a system that uses AI to implement a proven pedagogical framework, personalized to how each individual student actually learns.

The framework

Seven years of classroom data.
Encoded into a system.

The methodology came first. Seven years teaching GED prep — tracking what worked, what failed, and why — produced a set of principles about how learning actually happens. The AI is the tool that implements those principles at a level of personalization that was previously impossible.

Most AI in education is a chatbot that knows content. It answers questions. It explains concepts. That is useful — the same way a textbook is useful. But it is not teaching. Teaching requires knowing the specific student: what they understand, what they don't, how they think, where they get stuck, and what kind of explanation lands for them versus what bounces off.

The system I use builds that knowledge over time for every student, and uses it to drive every session that follows. The AI is not the teacher. It is the memory and the planner — the part that holds the student's full history and uses it to make every session smarter than the last.

"You are not getting a chatbot with a math degree. You are getting a teaching system built on a specific methodology, implemented with AI precision, and delivered by someone who has done this for real."

How it works

The four-stage
learning cycle

01

Diagnostic — reading first, then math

Every student starts with a TABE placement assessment and a 10-minute structured interview. The test reveals what they don't know. The interview reveals why — and what language, context, and explanatory style clicks for them. The reading baseline comes first, because reading comprehension predicts applied math performance more reliably than any other single factor. That result is documented and feeds directly into the Dynamic Learning Profile.

02

Dynamic Learning Profile — a living document

The diagnostic creates a structured profile for each student: current level by subject area, conceptual gaps, reading baseline, learning language preferences, and context anchors — the real-world framings that make abstract concepts concrete for this specific student. The profile is not static. It updates after every session. By session five, it knows more about how this student learns than most teachers discover in a semester.

03

The 80/20 calibration — productive struggle with a floor

Between sessions, the profile drives curriculum planning. Roughly 80% of each session targets the productive struggle zone — material at the edge of the student's current capability, where real growth happens. The remaining 20% reinforces mastery — material the student can execute with high accuracy. That 20% is not filler. It is the confidence infrastructure that makes the hard 80% sustainable. The calibration is updated after every session based on actual performance, not assumptions.

04

Session feedback loop — compounding precision

After every session, what worked, what didn't, and where the student is now gets fed back into the profile. The system gets more precise over time. A mistake in session one informs the approach in session five. The longer a student works within the system, the more accurately it targets their specific gaps — which is the opposite of how most tutoring works, where every session starts from scratch.

The methodology underneath

The principles the
AI implements

The AI does not generate the pedagogy. It executes a framework built from direct observation of what actually produces learning. These are the principles the system is built on.

Understanding over memorization

Methods that are understood transfer to new problems. Methods that are memorized collapse under novel conditions. Every explanation traces back to why.

Reading as the root of math

Most math errors in word problems are language errors. The diagnostic starts with reading because reading determines the math ceiling.

Productive struggle — calibrated

Growth happens at the edge of capability. But students who feel constantly lost disengage. The 80/20 split keeps the difficulty in the growth zone without breaking confidence.

Foundational fluency first

Cognitive load from slow arithmetic consumes the working memory needed for new concepts. Times tables, fraction rules, and order of operations must become automatic before higher-level work is reliable.

Performance vs. deep knowledge

Studying to pass a test and studying to understand are different modes that require different practice. The system switches between them deliberately.

Confidence as a variable

Confidence is not a personality trait — it is produced by specific conditions. The 20% mastery reinforcement in every session is confidence infrastructure, not padding.

Perseverance over IQ

Students fail because they stop engaging, not because they lack intelligence. The system is designed to keep students in the productive zone long enough for momentum to build.

Active recall over passive review

Session structure begins with retrieval before review. Memory is strengthened by the act of retrieval — not by rereading what is already familiar.

What's next

From methodology
to platform

The system currently runs through one-on-one sessions, with AI-assisted planning between them. The next step is encoding the full methodology into an interactive adaptive curriculum — a tool that implements the diagnostic, the 80/20 calibration, and the spaced repetition fluency drills automatically.

🎯

AI-Driven Diagnostics

Not just right/wrong — the system distinguishes reading errors from computation errors, conceptual gaps from fluency gaps. The diagnosis drives the prescription.

⚖️

The Calibration Engine

Automated 80/20 practice generation — every set of problems contains the right mix of growth-edge challenge and mastery reinforcement, adjusted in real time.

🔁

Spaced Repetition Fluency

The foundational fluency bottleneck — times tables, fraction rules, arithmetic — drilled with spaced repetition so cognitive load is freed for higher-level concepts.

📋

The Living Syllabus

A curriculum that updates based on the student's actual performance — not a static PDF, but a document that breathes and reacts to what the student demonstrates in real time.

⚙ Engineering Specification v3 → 📋 MVP 1 Tutor Manual → 📄 All white papers & research →

The methodology is live now in every session. First lesson free — experience it directly.

(702) 274-4299