Keep your teaching.
Gain adaptive learning.
Teacher-Controlled Adaptive Learning
AdaptEd takes the lesson you already teach, builds the checkpoints, records which sub-skill each student missed, and writes the follow-up for the students who need it. Nothing goes out until you say so.
Free tier: 40 pod generations a month, up to 30 students, 5 classes. No card, no expiry.
❯ The Choice You Shouldn't Have to Make
Today you pick one: keep control of your instruction, or get the automation.
Build it yourself
You keep every instructional decision.
- Your lessons, your sequence, your examples
- You write the assessment, then grade it
- Differentiation is a second set of materials
- The data lives in a spreadsheet you maintain
- Reteach, review, and documentation are all manual
Adopt a packaged program
You get the automation.
- Assignments, grading, and reports run themselves
- Intervention pathways are built in
- The vendor's lessons replace yours
- Its sequence, its explanations, its questions
- You manage someone else's system
Use AdaptEd
You keep the decisions. It does the rest.
- Your standards, your lesson, your material
- Checkpoints and exit assessments grade themselves
- Gaps recorded per student, down to the sub-skill
- Targeted follow-up — yours to approve before it goes out
- Missed content returns on a spaced schedule
❯ In Your Subject
Adaptive intervention has always stopped at K–8 reading and math.
Where the big platforms stop
Per their own published product materials: i-Ready's personalized instruction covers K–8 reading and math; Exact Path covers math, reading, and language arts; ALEKS and MATHia cover math. A validated item bank is built subject by subject, at cost — so nobody built one for high-school Biology.
Where AdaptEd runs
Subject is a free-text field, not a menu. Biology, Chemistry, World History, Computer Science, Financial Literacy — or whatever you actually teach. Nothing here depends on a prepackaged content library for your subject, because the lesson is generated from what you give it. Standards grounding is deepest where Texas TEKS exist for the subject; sub-skill and misconception data is richest in Biology.
❯ The Record It Builds
“Missed 9(B)” becomes “missed the light-dependent reactions.”
Sub-skill grain, not just a standard
Questions are tagged to a student expectation and to the specific sub-skill inside it — drawn from a taxonomy of 998 TEKS sub-skills. Your reteach targets the actual gap, not the whole standard.
It refuses to guess
When AdaptEd can't match a question to a sub-skill with confidence, it records that it couldn't — rather than guessing. An honest blank beats a wrong tag in a student's record.
Confidence intervals, not vibes
Mastery is reported with a 95% confidence range and an evidence gate. A lucky 5-for-5 doesn't read as “Mastered” here — we'd rather tell you the evidence is thin than tell you a student is fine.
Deliberately picky about evidence
From a lesson, only unaided first-try work counts — the solo run and the exit assessment. Guided practice and retakes are for learning, not for judging. Benchmark imports, spaced review, and auto-graded tool assignments join the same record. Everything still scores in your gradebook; this record sits alongside it.
❯ How It Works — One Week
One wrong answer becomes next week's targeted lesson. Every step below is yours to approve.
Monday — you teach
Enter a TEKS code, a topic, your own notes, or upload the lesson you already teach (PDF, DOCX, PPTX). AI drafts the slides, checkpoints, and exit assessment. It arrives as a draft only you can see — edit every slide, question, and answer key, then publish when you're satisfied.Tuesday — see exactly what broke
Students miss a checkpoint and get an immediate reteach and retry. Unaided first-try work lands in the mastery record by standard and sub-skill — not "61% on B.9B," but which nineteen students confused the thylakoid with the mitochondria.Tuesday, 3:47pm — you decide who gets reteaching
Open Standards Mastery: every student below the Meets floor is already checked, and so is every sub-skill the class is weak on. Untick anyone you disagree with. One click builds one targeted reteach for that group, and hands it back for your review.Every day after — it comes back until it sticks
Missed questions return in two-minute daily reviews on a spaced schedule — 1, 3, 7, 14, then 30 days. Miss it again and it goes back to the start of the cycle, so it keeps returning until they get it right all the way through.❯ For Teachers
Teach your way. Hand off the repetition.
Start from what you already have
A TEKS code, a topic, your own notes, a PDF, a Word doc, a slide deck, a YouTube video, questions from your own bank, or a colleague's pod from the community library. Scanned handouts run through OCR. Your material is the primary source the AI teaches from.
Nothing reaches students unedited
Generated lessons arrive as drafts only you can see. Rewrite any slide, bullet, or speaker note; rewrite any question stem and its answer key across every activity type; add, delete, or regenerate. A pre-publish check flags anything a student couldn't actually answer.
Grading that doesn't wait for you
Ten interactive question types — matching, sorting, drag-into-passage, image hotspots, categorization, two-part multiple choice and more — are graded server-side by deterministic rules, not by an AI guess. Code, SQL, and spreadsheet responses are AI-graded on the Pro tier.
Differentiate without building it twice
Reteach a whole group from one screen, or generate a make-up pod for a single student aimed at their weakest sub-skills and the specific misconceptions they showed. Preview the exact evidence the AI will use — and uncheck any of it — before it generates.
See what each student needs next
Standards Mastery ranks every student weakest-first on any standard you've assigned, with their band, attempt count, confidence interval, and unresolved misconceptions — plus the class average on each sub-skill inside that standard.
Stop re-entering the same work
Google Classroom rosters sync in and grades sync out. Benchmark and STAAR-style CSV results import straight into the same student record. Assignable practice built from seven of the quick tools grades itself and posts back.
❯ What's Inside a Pod
Not a worksheet. Not a video. A complete learning experience.
Interactive Slides
AI-generated slides with embedded media and guided notes. Dwell-time tracking means clicking straight through doesn't go unnoticed.
Built-In Checkpoints
Matching, sorting, drag-into-passage, image hotspots — STAAR-style practice in the lesson flow, plus AI-graded code, SQL, and spreadsheet answers on the Pro tier. Wrong answers trigger immediate reteach.
Built to Be Cheat-Resistant
Locked exit assessments, mastery gates, focus mode, and dwell-time and rapid-click signals give you a timestamped, auditable integrity record — so you act on the pattern instead of guessing. No web tool can stop a student photographing a screen, and we won't claim otherwise.
Spaced Repetition
Missed concepts resurface in daily review micro-pods on a Leitner schedule. Mastery is proven over time, not claimed by one score.
❯ For Schools
No adoption cycle, no IT project, no district sync.
Teachers keep their own lessons, their own sequence, and their own materials. You get standards-level data across every classroom that connects — no scripted curriculum, no adoption cycle, no rollout project.
Teacher-authorized, read-only access
No district sync, no IT project: you issue a campus code, each teacher enters it once, and they can disconnect in one click. Read-only on your side, and no click-through into an individual student's answers — that stays with their teacher.
Standards-level data by class and teacher
The standards your campus is weakest on, with campus average, students assessed, and attempts — plus a performance-band distribution across every connected class. A student needs five or more scored attempts before any band is asserted, and thin evidence is reported as thin rather than as a verdict.
❯ The Rest of the Week
Supporting the loop — not the point of it.
Plan
Weekly lesson plans and station rotations, with your standards attached.
Teach
Lesson hooks, differentiated reading, and word walls.
Practice
Pods live or self-paced, five hands-on science simulators, and daily review.
Assess
Locked exit assessments, a question bank with OCR import, and benchmark CSV import.
Respond
Standards Mastery, misconception tracking, reteach pods, and spaced review.
Science Simulators
Membrane transport, Punnett squares, enzymes, mutations, cell builder — students experiment, and the result is graded by hand-written rules rather than an AI opinion.
Guardrailed AI Tutor
Stuck students get a targeted nudge, not the answer. Help escalates from concept hints to partial steps while the student writes the final response — with daily caps, and an off switch per pod.
Google Classroom
Rosters sync in, coursework and grades sync out, with per-student push state. Seven tools also export to Google Forms, four to Docs.
Assignable, Auto-Graded Practice
Seven of the quick tools — word walls, card sorts, task cards, study cards, Color-by-Number, differentiated reading, source lessons — can be assigned digitally, grade themselves, and update the student's standards record.
❯ Quick Start
Pick your seat — the loop looks different from each one.
❯ Why It's Built This Way
Good teaching runs on judgement, relationships, and knowing the room — none of which software has. What software is good at is the part that doesn't fit in a planning period: grading a hundred exit tickets, remembering which sub-skill each student missed in October, and having the follow-up ready before Thursday.
So AdaptEd never decides what should be taught. It takes your decision and does the bookkeeping around it — and when it isn't sure, it says so instead of guessing.
❯ What We Can and Can't Tell You Yet
The honest version, because you'll find out anyway.
Built by a practicing Texas teacher
Not a startup's guess at what a classroom needs. Every workflow here exists because the person who wrote it needed it on a Tuesday.
A glass box, not a black box
You can see every answer every student gave, on every question. Scoring for the ten interactive question types is deterministic and server-side — the same input always produces the same verdict. No proprietary score you're asked to trust.
No third-party tracking
No analytics suite, no advertising pixel, no session recorder — none, anywhere on this site or in the product. Student data supports a self-service data request and deletion path.
No efficacy claim — yet
We have no independent study, no test-score guarantee, and no case studies to show you. A controlled classroom pilot runs this fall. When there's a number, it will appear here with its method attached; until then, judge the product on what it does.
Keep control of your instruction. Add the adaptive part.
Know which students need help before the benchmark lands — and have the follow-up drafted, waiting for your approval, and coming back on a spaced schedule before you leave for the day.