When your brain works differently, AI isn’t a luxury—it’s accessibility
AI serves as a critical accessibility tool for neurodivergent professionals, specifically compensating for executive function gaps associated with AuDHD (Autism and ADHD co-occurrence). The author built a self-maintaining AI workflow system using Amazon Quick, which automates email triage, task prioritization, and context management to reduce cognitive load. The architecture leverages the Model Context Protocol (MCP) server to connect AI assistants with enterprise tools like Outlook and Asana, a
Analysis
TL;DR
- AI serves as a critical accessibility tool for neurodivergent professionals, specifically compensating for executive function gaps associated with AuDHD (Autism and ADHD co-occurrence).
- The author built a self-maintaining AI workflow system using Amazon Quick, which automates email triage, task prioritization, and context management to reduce cognitive load.
- The architecture leverages the Model Context Protocol (MCP) server to connect AI assistants with enterprise tools like Outlook and Asana, allowing for rule-based automation via configurable markdown files.
- The system utilizes Amazon Bedrock for inference and Kiro IDE for development, enabling adaptive reasoning without requiring code redeployment when rules change.
- Key design principle focuses on minimizing the cognitive cost of usage, shifting the burden of organization from the user to an automated, deterministic background process.
Why It Matters
This case study highlights a significant shift in AI utility from general productivity enhancement to specialized accessibility support, addressing the needs of the estimated 15–20% of the population that is neurodivergent. It demonstrates how integrating AI into existing workflows can mitigate the unique cognitive challenges of conditions like AuDHD, such as decision paralysis and working memory deficits. For the broader industry, it underscores the importance of designing AI tools that reduce, rather than increase, the executive function required to manage them.
Technical Details
- Core Architecture: The system is built on Amazon Quick, an AI-powered desktop assistant that provides persistent memory and tool orchestration, connected to Amazon Bedrock for underlying LLM inference.
- Integration Layer: A custom Model Context Protocol (MCP) server acts as the bridge, connecting the AI assistant to external applications including Outlook (email/calendar) and Asana (task management).
- Configuration Management: Triage rules, priority logic, and communication patterns are encoded in configurable markdown files. The MCP server reads these files fresh each session, allowing immediate behavioral updates without redeployment.
- Development Environment: The MCP server was developed using Kiro, an AI-powered IDE from AWS, facilitating rapid iteration and integration.
- Automation Framework: The "Quick skills" framework provides reusable, deterministic automation patterns for recurring tasks such as email formatting, context logging, and end-of-day summaries, ensuring consistent execution with minimal user intervention.
Industry Insight
- Accessibility by Design: AI product teams should consider neurodiversity in UX design, focusing on reducing executive function friction. Tools that automate organization and prioritization can unlock productivity for a significant portion of the workforce.
- Low-Maintenance Automation: The success of this system relies on its ability to run autonomously after a single initiation step. Future AI tools should prioritize "set-and-forget" capabilities that adapt to user preferences without requiring constant manual oversight or complex setup.
- Standardized Integration Protocols: The use of MCP demonstrates the value of standardized protocols for connecting AI agents with enterprise software. This approach allows for modular, rule-based customization that can evolve independently of the underlying model improvements.
Disclaimer: The above content is generated by AI and is for reference only.