Bounty Scoring
How submissions are evaluated
Scoring System
How we rank submissions. Fair, transparent, merit-based.
TLDR
- AI-Powered Evaluation — Every submission runs through AI to check alignment and authenticity
- Quality Gates — Must align with bounty brief + can't be AI-slop. We block misaligned submissions.
- Format Verification — Video bounty? We check it's a video. Thread bounty? We check it's a thread.
- Weighted Engagement — Not all likes are equal. Verified accounts, public profiles, and respected industry voices count more.
- 0-1 Scoring — Every metric (likes, comments, RTs) scored 0-1. Zero = noise, One = signal.
- Fair Distribution — Higher scores = bigger share of the prize pool
What it is
First Dollar uses AI-powered evaluation combined with weighted engagement data to score bounty submissions. We don't just count likes—we measure real impact.
Why it matters
No more fake engagement or AI-generated spam winning bounties. Every submission gets evaluated on quality AND authentic performance. Ship great work, get real engagement, earn your share.
AI Evaluation
Every submission runs through our AI system. We check three things:
1. Content Alignment
Is the submission actually about what the bounty asked for?
- Does it match the bounty description?
- Does it hit the required topics?
- Does it follow the brief?
If it's not aligned, it gets blocked. No exceptions.
2. Format Verification
Is the submission in the right format?
- Video bounty? We verify it's actually a video
- Thread bounty? We verify it's actually a thread
- Single tweet? We check it's not a thread when it shouldn't be
Wrong format = blocked. Read the brief.
3. AI Detection
Is this authentic human work or AI-generated slop?
- We detect heavy AI usage
- We flag copy-paste submissions
- We check for authentic voice and perspective
If it's clearly AI-generated, it gets blocked. We want real creators, not prompt engineers.
Weighted Engagement (0-1 Scoring)
Not all engagement is equal. Every interaction gets scored from 0 to 1:
- 0 = Noise (bots, fake accounts, irrelevant)
- 1 = Signal (real people, relevant audience, meaningful interaction)
What We Track
For every like, comment, retweet, and reply, we check:
| Factor | Why It Matters |
|---|---|
| Verified Account | Verified users have more weight than unverified |
| Public vs Private | Public accounts count more than private profiles |
| Industry Voice | Known builders and popular, relevant accounts score higher |
| Account Quality | Real accounts with history vs. fresh/suspicious accounts |
How It Works
A like from a verified builder with 50k followers counts way more than 100 likes from fresh accounts with no posts.
We're measuring real reach to real people, not vanity metrics.
Engagement Metrics
For Twitter/X bounties, we track:
- Views — How many people saw the content
- Likes — Audience appreciation (weighted by account quality)
- Retweets — Amplification and reach (weighted)
- Quote Tweets — Engagement with commentary (weighted)
- Replies — Conversation generated (weighted + AI-scored for relevance)
Each metric gets weighted and combined into a Combined Total Engagement score.
Leaderboard
The submissions tab shows a live leaderboard with:
- Submission rank
- Builder name
- Tweet link
- Individual metrics (likes, views, RTs, quotes, replies)
- Combined weighted score
Leaderboard refreshes every 6 hours to capture ongoing engagement.
"Rankings are provisional. Submissions may be removed if they don't meet the bounty requirements."
Prize Distribution
For Contest Bounties
You pick the winners manually. Scores help you compare, but final decision is yours.
For Prize Pool Bounties
The prize pool gets split automatically based on weighted engagement:
- Your share = Your weighted score ÷ Total weighted scores × Prize Pool
- Top performers get proportionally more
- Quality engagement beats quantity every time
What Gets Blocked
Submissions are automatically blocked or removed if they:
- Don't align with the bounty requirements
- Wrong format (video when it should be thread, etc.)
- Heavy AI usage without authentic input
- Fake engagement from bots or purchased likes
- Spam or low-effort copy-paste content
- Platform violations or reported as spam
Always Improving
We're constantly refining the scoring system to make it fairer and more accurate. Expect improvements to roll out with upcoming bounties—better AI detection, smarter engagement weighting, and more transparent scoring.
The scoring system ensures fair rewards for quality content from real creators. No bots. No slop. Just builders earning their first dollar.