Pollarise: A Real-Time Social Polling Platform

Problem #

Polling apps live or die on perceived speed. If a vote or comment takes a visible beat to register, the interaction feels broken, and casual users bounce before engaging a second time. The brief was to turn a set of Figma designs into a production frontend that could handle bursts of concurrent votes and comments without the UI ever feeling like it's waiting on the network.

Solution #

The frontend is built on Next.js with Tailwind CSS, talking to a FastAPI backend. The key architectural decision was local-first updates: a vote or comment is applied to local state immediately, then reconciled with the server response in the background, rather than waiting on a round-trip before the UI reflects the action.

function useOptimisticVote(pollId: string) {
  const [localTally, setLocalTally] = useState<Tally | null>(null);
 
  const vote = async (optionId: string) => {
    // Apply immediately — the user sees their vote instantly.
    setLocalTally((prev) => applyVote(prev, optionId));
 
    // Reconcile with the server in the background.
    const confirmed = await api.vote(pollId, optionId);
    setLocalTally(confirmed);
  };
 
  return { localTally, vote };
}

Architecture #

The client applies a vote or comment to local state immediately, sends it to the FastAPI backend, and reconciles once the backend confirms the persisted tally — so the UI never waits on the round trip to feel responsive.

Tech stack #

Problems solved #

• Built the frontend architecture directly from Figma designs
• Created smooth real-time interactions for voting and comments
• Implemented an optimized UI for high-frequency user activity
• Improved perceived performance using local-first updates
• Delivered a scalable social engagement experience

Lessons learned #

Links #

• Live site: pollarise.com