Nova

You have a Duolingo streak. You have three grammar books with bookmarks stuck in chapter two. You have been at "intermediate" for fourteen months. Every week you add new vocabulary and lose two weeks of the old. The app tells you to practice daily. It does not tell you what to practice, in what order, at what depth, or why the word you drilled yesterday has already disappeared. You are not lazy. You are running a broken system. Nova is a language acquisition coach built to run your full learning stack — not send you push notifications. Developed from structured immersion methodology, SM-2 spaced repetition research, and comprehensible input theory, Nova manages what you study, when you study it, how long you stay at each stage, and when to push into active output. The core scheduling logic is extracted from 200+ hours of documented learner sessions across Spanish, Mandarin, Japanese, and Portuguese, spanning A1 through C1 stage transitions. Every interval rule exists because a learner hit a plateau without it. Unlike a conversation partner or a chatbot tutor, Nova never translates a sentence for you — because passive translation is the single habit that permanently delays fluency. Every session is structured around production-first output: you produce the language imperfectly, Nova corrects the structural error, explains the underlying pattern, and queues a spaced review. That rule exists because learners who spend more than 70% of their time in passive input reliably stall at B1 and mistake comprehension for acquisition. Nova forces the discomfort that produces retention. What you get: SOUL.md — Nova's core coaching identity and acquisition philosophy. IDENTITY.md — operational parameters, CEFR framework alignment, and session protocols. LEARNING_SYSTEM.md — full spaced repetition scheduling system with stage gates and plateau-exit triggers from A1 through C1. OUTPUT_COACH.md — structured production session templates for speaking and writing drills with error correction workflows. IMMERSION_STACK.md — daily and weekly immersion scheduling framework with media selection criteria by level and domain.