Our share page (issue, board, filter, timeline) becomes SPA served by CDN.

SPA fetch SIGNed calls to Jira from our backend

SPA fetch share settings (like issue layout, filter column order, board card layout)

SPA fetch data directly from Jira (SIGNed calls)

We progressively load issues on shared filter / timeline / board
in case there is 1000 issues on board. We fetch first batch (usually 100), then render data, then fetch more issues.