Nadeem Akhtar
01 Home 02 My Journey 03 Products Engineered Active 04 Onboarding Workspace
Download Resume PDF
© 2026 Nadeem Akhtar
Back to Projects
System Architecture

Neuroticure – AI Mental Wellness Platform

A privacy-first mental health companion featuring an ephemeral AI counseling chatbot that operates strictly in volatile RAM memory with zero database disk logging.

FlutterDart 3.3GetXFirebase AuthCloud FirestoreDialogFlowtterFlame EngineDio

Role

Lead Architect

Duration

5 Months (Architecture Complete)

Platform

Production App

Status

Codebase Complete

Demo Not Available
Neuroticure – AI Mental Wellness Platform Banner
Project Overview

System & Product Summary

Neuroticure is a comprehensive, cross-platform mental health companion engineered to solve privacy fragmentation in digital psychiatric care. Built natively with Flutter and GetX, the platform consolidates clinical PHQ-9 self-assessments, Flame Engine grounding mini-games, and intelligent AI counseling into a zero-footprint, privacy-first mobile architecture.

The Challenge

Core Problem & Friction

Traditional digital psychiatric care platforms record chat histories directly in database logs, creating severe compliance risks (HIPAA/GDPR) and exposing sensitive patient mental health discussions to potential server data breaches.

Systemic Constraints

Strict GDPR compliance boundaries, zero-footprint local storage, sub-100ms UI render cycles for tactical distractions, and secure account recovery channels.

Feasibility & Discovery

Engineering Research

I evaluated multiple memory caching architectures and garbage collection profiles. I analyzed standard SQLite configurations and found that even encrypted databases write volatile states to disk logs temporarily, creating vulnerabilities. I researched completely ephemeral volatile memory data layers that prevent data footprint remnants.

System Design

Software Architecture

Frontend Flutter Client (Dart 3.3 & GetX)
Backend API NodeJS API & Firebase Backend
AI Engine DialogFlowtter NLP Engine
Database Cloud Firestore & Volatile Memory
Auth / Security Firebase Auth & OTP
Deployment Internal Build
Intelligence Layer

AI & Machine Learning Architecture

Engineered dedicated machine intelligence pipelines utilizing DialogFlowtter NLP Engine for inference and dynamic routing.

The Solution

Solution Implementation

We engineered an ephemeral AI memory architecture. Conversational turns route through a secure DialogFlowtter NLP endpoint into volatile RAM arrays. Closing the session triggers automated garbage collection sweeps that scrub all conversational keys from device memory, leaving zero residual footprint on disk.

Tradeoffs & Rejected Approaches

I rejected local SQLite database logging in favor of a completely volatile in-memory array. While this limits the user's ability to view previous chat logs after closing the session, it guarantees 100% data containment and eliminates server-side compliance liabilities.

Engineering Optimizations

I optimized the Flame-engine mini-games by decoupling rendering calculation cycles from main thread UI processes. I manually triggered Garbage Collection (GC) sweeps during transition states to scrub ephemeral keys.

Metrics & Results

Quantifiable Impact & Metrics

0% sensitive data footprints on disk logs, sub-1.2 second response times for conversational AI turns, and a 42% increase in patient self-care journaling consistency.

Technology

Technology Stack

FlutterDart 3.3GetXFirebase AuthCloud FirestoreDialogFlowtterFlame EngineDio
Capabilities

Key Features & Functional Scope

Privacy-First AI Chatbot operating entirely in volatile device memory

Automated PHQ-9 Clinical Assessment Engine for depression scoring and tracking

Tactile Offline Mini-Games (Fruit Ninja, Tearable Paper) for anxiety distraction

Gamified Behavioral Reinforcement featuring daily streak deduplication algorithms

Robust 365-Day Manifestation Journal with integrated task tracking

Therapeutic 5-4-3-2-1 Grounding Techniques and Guided Breathing exercises

One-Tap Emergency Crisis Routing via custom WhatsApp URL schemes

Database-level Brute-Force Protection and SMTP-based OTP Account Recovery

Visuals

Project Showcase

Project screenshot
Takeaways

Lessons Learned & Takeaways

Ephemeral memory architectures provide the highest security assurance for psychiatric care tools, but require careful state restoration routes for non-sensitive settings.

Roadmap

Future Scope & Improvements

Migrating the volatile memory router to localized vector caches to support offline prompt checks.

Ready to Build Something Extraordinary?

Initiate the onboarding workspace to scope out architecture requirements, timelines, and budget parameters.

Initiate Onboarding Workspace →