Filtering large datasets through a query DSL

A filter system that compiles UI state into a central query service’s DSL, with a new distinct-values endpoint, typeahead over very large sets, keyset pagination and persisted state.

When: Jul 2026
Role: Engineer
Context: Explorer, the account drill-down surface

Context

Explorer is Optmyzr’s account drill-down surface, being extended to Meta and LinkedIn Ads. It needed a global filter bar from a new design — one that works on accounts with very large campaign sets.

What I built

A filter-bar component library and a Redux slice with Firebase-persisted state.
Compilation of the filter state into the central query service’s filter DSL.
A new backend operation that returns complete distinct values per attribute (a grouped query), so pickers show every real option rather than whatever happened to be on the current page.
Campaign-name typeahead with capped, virtualized options.
Saved-views integration and strings in five languages.
Server-side search by entity name or ID, keyset pagination (including a fix for a non-advancing cursor), and killing a runaway request loop.

Outcome

Built in a two-day burst of about 25 scoped commits as part of the Meta & LinkedIn Explorer epic, and part of the Explorer customer beta that opened in September 2026.