Merilin Ekzarkova

I write, I edit, and I build intelligent systems.

Product lead. I evaluate existing systems, identify the problem, and use AI tools to design and build an independent, self-sustaining system that solves it.

I look at how things work now, find what's broken, and use AI to build something that fixes it and keeps working without me.

Work

Built a live dashboard tracker for client trials, kept up to date by an AI agent

Problem
Trial data sat in four systems and was brought together only once a trial had ended, too late to act on during it.
Solution
A daily picture of engagement and retention from 4 data sources. In its first week it showed who had not come back and flagged two data problems that would have reached the final report.
Read more →
Sketch of a trial tracker: a score out of 100, a bar for each day, a weekly report and four data sources 72 score out of 100 active members, day by day Weekly report Approve Sign-upsApp useSurveysChat
Sketch with made-up numbers, not the real screen.

Built a portal and tracker for partner feedback and QA

Problem
Partner feedback arrived as slide decks. Each one needed an hour-long team meeting, and there was no log to show what had been asked before.
Solution
One online portal and a log of every request. A round now takes one person about 15 minutes, and repeats are spotted at once.
Read more →
Sketch of a feedback portal: a form to submit an issue, and a log of how often each change was requested New issue Type screenshot Submit Request logwhat, how often, when Wordingasked 3×Layoutasked 1×Wordingasked 2×
Sketch with made-up numbers, not the real screen.

Audited 49 app screens for accessibility and shipped the fixes

Problem
The app had no documented accessibility baseline ahead of a planned design refresh.
Solution
A 49-screen audit against WCAG 2.1 AA, turned into 7 workstreams. 6 shipped within the year, as 15 changes in the app.
Read more →
Sketch of a phone screen marked up with accessibility checks: a description for a picture, a button measured at 44 points, and a progress label Continue a description,read aloud step 1 of 3 44 points:room for a thumb
Sketch of the kinds of checks, not a real screen.

Also: a decision-support dashboard for public-sector health teams, app delivery of about 210 changes in 11 months, a terminology system for a Japanese release, and the analytics tracking plan. See all work.

Writing

Also in UX Collective: The preventive healthcare product cycle. More essays are on my Medium page.