Built by educators and engineers who believed the problem was worth solving.

Jessica Tanaka

CEO & Co-Founder

Jessica spent 7 years teaching 7th and 8th grade STEM in the Portland school district before founding Brainpathio. During that time she maintained a hand-written taxonomy of student misconceptions — 60+ categories she discovered were repeating year after year across different students.

The insight that became Brainpathio came from a single observation: two students with the same wrong answer on the same test needed completely different explanations. One needed 5 minutes. The other needed two weeks. The test score told her nothing about which was which.

Marcus Osei

CTO & Co-Founder

Marcus met Jessica at a Portland tech community event in early 2024 where she was walking through her hand-written misconception taxonomy. He told her the problem she was describing was a Bayesian inference problem — a probability distribution over competing hypotheses, updated with each student response. She told him to prove it. He did.

Over a 6-week sprint in summer 2024, Marcus built the first version of the hypothesis engine: the core system that takes a student's response pattern and maintains a running probability distribution over misconception categories, updating it with each answer. The first validated classification result was a 7th-grader's "variable as label" misconception that Jessica had flagged manually from a teacher-friend's classroom. The engine matched her judgment with 89% confidence.

Priya Narayan

Head of Curriculum Science

Priya joined Brainpathio from a curriculum research role at a K-12 education policy organization, where she had spent four years evaluating the evidence base for mathematics intervention programs. Her job at Brainpathio is to connect Jessica's classroom-derived taxonomy to the peer-reviewed literature — verifying that each misconception category the engine tracks has documented research backing, from Küchemann's variable confusion work to Clement's Aristotelian motion studies.

She also owns standards alignment: mapping every misconception category explicitly to the CCSS-M domain or NGSS disciplinary core idea it blocks. "If a teacher can't show their department head which standard the alert connects to, the alert is useless," she says. "The whole taxonomy has to be auditable — not just plausible."