A privacy-first iOS app that helps neurodivergent individuals track behavioural patterns, recognise triggers, and gain meaningful insights. Your journal lives on your device, and when AI features need processing, data travels encrypted to EU servers and is never stored.
Standard habit-tracking apps are built on compliance, daily check-ins, streaks, and prompts. For individuals with Pathological Demand Avoidance, each micro-demand triggers an avoidance response, leading to permanent abandonment of the tool.
During autistic burnout, the cognitive overhead of opening an app, navigating menus, and filling structured forms is physiologically impossible. Existing tools require precisely the resources that burnout depletes.
Current apps apply a fixed interaction model regardless of a user's daily neural capacity. A neurodivergent individual's capacity fluctuates significantly, no tracker adapts its interface demands to this reality.
Meet Mimo
Mimo reads everything you've written. It doesn't give generic advice, it answers from your own history and pushes back when you're jumping to conclusions.
Speak naturally. On-device transcription understands medication names, emotional shorthand, and neurodivergent speech patterns that standard dictation gets wrong. Or type, whatever takes less energy.
Every entry is classified into one of six types: observation, venting, reflection, insight, strategy, or positive. The AI understands what you're doing, processing, reflecting, coping, and organizes accordingly.
An AI assistant that has read everything you've ever written. Ask "what strategies have worked for my fatigue?" and Mimo searches your history, retrieves the evidence, and answers from your own data, not generic advice.
Day view shows each entry with type-aware context. Week view clusters entries into threads, what you're working through. Month view reveals arcs across weeks. Each level answers a different question.
Sleep, heart rate, HRV, and workouts from Apple Health, correlated with your entries to give the AI real physiological context alongside your subjective experience.
Medications, supplements, and accommodations organized by time of day. Log doses with a tap, track what you've taken, and see your routine at a glance, no spreadsheet required.
Cortex uses large language model architecture as a conversational, low-demand interface. Rather than forcing structured forms, the system extracts behavioural data from natural ambient input, voice fragments, short text, or passive signals.
The backend runs on AWS Bedrock using encrypted EU servers with zero data retention, fully GDPR compliant. Local testing uses Ollama, enabling rapid iteration without cloud dependency.
The architecture is model-agnostic: the same behavioural extraction layer works across hosted foundation models and on-device inference, so the system degrades gracefully in low-connectivity environments typical of burnout states.
How Your Data Flows
Your journal stays on your device
Explored how AI can support neurodivergent self-tracking. Validated the core concept: free-form journaling with automatic classification and pattern extraction. Defined the architecture and data model.
Building the iOS prototype with on-device voice transcription, AI-powered classification, and HealthKit integration. Testing with personal use to refine the experience before wider release.
Public beta on TestFlight for early testers from the neurodivergent community. Gathering feedback, refining the AI, and preparing for App Store submission.
Two Perspectives
Natural shows how Cortex feels in everyday life, warm, intuitive, human.
Cortex shows how your brain actually works, through the language of neuroscience.
Traditional habit trackers fail when executive dysfunction hits. Cortex uses ambient AI to map Pathological Demand Avoidance (PDA) and autistic burnout without requiring manual data entry.
Join the 2026 Beta WaitlistSystem Failures
Standard habit-tracking applications are built on a foundation of compliance, daily check-ins, streaks, and prompts. For individuals with Pathological Demand Avoidance, each of these micro-demands triggers an autonomic avoidance response, causing immediate and permanent abandonment of the tool.
When users are experiencing autistic burnout, the cognitive overhead required to open an application, navigate menus, and fill structured input forms is physiologically impossible. Existing tools require precisely the neurological resources that burnout depletes.
Current applications apply a fixed interaction model regardless of the user's daily neural capacity. A neurodivergent individual's functional capacity fluctuates significantly across hours and days. No commercially available tracker adapts its interface demands to this reality.
Mimo RAG Engine
Mimo queries your full journal history via hybrid search (vector similarity + FTS5), then interprets from raw context. It cross-references timestamps, physiological data, and prior entries to challenge premature conclusions.
On-device speech-to-text via WhisperKit. No cloud roundtrip for transcription. Neurodivergent speech patterns parsed locally, then structured by the extraction pipeline.
Every input classified: observation, venting, reflection, insight, strategy, positive. Each type triggers different extraction rules and contextual rendering.
Retrieval-Augmented Generation over full journal history. Hybrid search, vector similarity + FTS5 keyword, retrieves relevant entries. LLM interprets fresh from raw text every time.
Weekly and monthly summaries are generated automatically from the extracted behavioural data, providing visual, structured output of trends and pattern classifications.
Entries clustered by topic using primaryEntity and semantic similarity. Weekly threads show what you're working through. Monthly arcs reveal persistence across weeks.
Medications, supplements, accommodations tracked with time-block scheduling. Multi-dose support, HealthKit workout auto-sync, validation status lifecycle.
Cortex utilises advanced large language model (LLM) architecture to serve as a conversational, low-demand interface. Rather than forcing the user to interact with structured forms, the system extracts behavioural data seamlessly from natural, ambient input, voice fragments, short text, or passive signals.
The backend inference pipeline is built on AWS Bedrock EU (Frankfurt), providing enterprise-grade model access with GDPR Article 28 compliance and zero data retention. During the active proof-of-concept phase, local model testing is conducted via Ollama, enabling rapid iteration without cloud dependency or data egress.
The architecture is model-agnostic: the same behavioural extraction layer operates across hosted foundation models and on-device inference, ensuring the system degrades gracefully in low-connectivity environments typical of burnout states.
Technical Architecture Diagram
To be replaced with final architecture illustration
Validated that unstructured journal input can be parsed into structured behavioural data. Defined extraction pipeline, classification taxonomy, and on-device storage architecture.
Building iOS prototype with cloud coprocessor (AWS Bedrock, EU endpoints). On-device voice, semantic search, 6-type classification, RAG chat. Testing with personal use.
TestFlight beta for neurodivergent early testers. Gathering feedback before App Store submission.
Two Perspectives
Natural shows how Cortex feels in everyday life, warm, intuitive, human.
Cortex maps your experience to how your brain actually works, through the language of neuroscience.