As a seasoned developer, my day involves looking at a lot of screens. But the most exhausting part of my job isn't writing C# or debugging JavaScript-it’s navigating the project management tools we are forced to use.
Modern productivity tools (Jira, DevOps, Asana) have a massive accessibility problem that nobody talks about: The Wall of Text. If you have dyslexia, or any kind of neurodivergence that affects how you process information, opening a standard Agile dashboard is an instant cognitive drain.
Scanning through rows of tiny, low-contrast text, trying to differentiate between ticket DEV-405 and DEV-450, and digging through nested menus just to update a status requires a huge amount of mental energy.
I needed a tool that was visually quiet. Something that respected my brain's processing limits.
So, I stepped outside my usual stack, learned Python, and built SheepCat-TrackingMyWork an open-source, local AI task tracker designed for neuro-inclusive accessibility.
If you look at the UI, you'll notice it goes completely against the modern trend of cramming as much data on screen as possible.
Here is how being dyslexic shaped the design:
1. High Contrast & Big Targets
Look at the Tracker Page. There is no dashboard to read. There are three massive, color-coded buttons with clear, bold text. When the app gently prompts you for an update, your brain doesn't have to decode a complex menu. You know exactly where to click. Visual hierarchy is prioritized over text density.
2. Flat Tables, No Nesting
On the Review Work Log page, the table is heavily spaced with distinct row highlighting. You can double-click to toggle a "Resolved" status without having to open a text-heavy modal or read through a dropdown menu. It reduces the reading load to absolute zero.
3. The Secret Weapon: AI as a Writing Buffer
One of the hidden struggles of dyslexia in tech is "spelling anxiety"-the stress of writing professional, typo-free updates in public Jira tickets or Slack channels.
SheepCat uses interstitial logging. It prompts you hourly, and you can just dump your raw, messy, misspelled thoughts into the box. At the end of the day, the app feeds those raw CSV logs into a 100% local Ollama model (keeping enterprise data completely private). The local AI pieces together your messy notes and writes a clean, polished daily stand-up summary for you.
The Takeaway
When we design tools to be accessible for dyslexic and neurodivergent brains, we usually end up creating a UI that is less exhausting for everyone.
I’ve open-sourced the whole project under the GNU AGPLv3. You can check out the Python/Docker setup, and my UI style guide, on GitHub here or check out our website
Top comments (4)
Awesome perspective and tool! From a technical standpoint, the task tracker approach makes a lot of sense — breaking work down into structured, modular units reduces cognitive load and aligns with how developers naturally reason about problems. By minimizing dense blocks of text and focusing on actionable items, you’re essentially applying principles of information chunking, task decomposition, and cognitive ergonomics that improve clarity and maintainability in both personal workflows and codebases. Great example of designing tooling that matches how people think, not just how systems are built
Thank you! 'Cognitive ergonomics' is a brilliant way to phrase it. Truthfully, it just started as a tool that did exactly what I needed to survive the workday. The local AI was a massive easy win (especially on the spelling side 😅). I'm just thrilled to see that building something for my own brain might actually be useful to others too. Thanks for checking it out!
I feel seen thanks for this
Knowing it made someone else feel seen is the best feedback I could ask for. Thank you for taking the time to let me know ✨