We Built the K-2 Math Screener — Because Student Achievement Needed It
The data on student achievement is clear. What is less understood is why the gap persists — and what can specifically be done about it at the classroom level.
Brown Forces spent time working in this space. We talked to teachers, we studied the assessments that currently exist, and we found a problem inside the problem.
**The misdiagnosis**
Standard math screeners cannot distinguish between two very different students: one who genuinely does not understand place value, and one who understands place value completely but cannot decode an English word problem. Both score the same on a standard screener. Both get labeled as having a math gap. One of them does not.
In schools serving English Learner communities — which describes most of the schools in the Coachella Valley, the Central Valley, South Texas, and across the country — this misdiagnosis is not an edge case. It is the rule. Students who are mathematically capable are routed into remediation programs designed for students with content gaps. The intervention does not match the need. The gap does not close.
We decided to build the tool that makes the distinction.
**What we built**
The Brown Forces K-2 Math Screener is an AI-adaptive assessment platform for kindergarten through 2nd grade, aligned to California Common Core Math standards — 69 standards, 61 bilingual questions covering K.CC through 2.G.
Claude Haiku powers the adaptive engine. After each student response, the AI analyzes the full response history and selects the next question: advance to a harder concept, probe a prerequisite, or trigger a language gap probe.
The language gap probe is the core innovation. When a student fails a word problem, the engine immediately follows with a pure computation version of the same concept. Fails the word problem, passes the computation — that is a language gap, not a math gap. The system records it as such and surfaces it to the teacher differently from a content gap.
**What the teacher sees**
A gap map for every student. Standards color-coded by mastery level. Language gaps rendered distinctly — so the teacher knows immediately which students need math intervention and which students need language support. Those are different instructional decisions and they should never be conflated.
A principal view surfaces the school-wide language gap rate by grade — what percentage of students who appear behind in math are actually experiencing a language barrier. That metric has direct policy implications, particularly in the context of California's ongoing focus on English Learner outcomes.
**How it is built**
FERPA-compliant. No student names are stored — only anonymous IDs assigned by the teacher. Fully bilingual English and Spanish. Optimized for iPad. No student login required.
The screener is live at screener.brownforces.io. Any teacher can use it today.
**Why we built it first**
We built the tool before anything else because we wanted educators to see it work, not just read about it.
Districts, schools, and teachers who want to explore it can reach us through the contact page. We are ready to support implementation, training, and whatever comes next — on the district's terms and timeline.
The tool is live. The work continues.