Inspiration

Some of us on the team have been creators ourselves, from posting short-form videos to running small online communities. We know firsthand the feeling of spending hours on a piece of content, only to watch it sink while a low-effort trend goes viral. These experiences sparked our question: what if creators could see exactly how their content is being judged, scored, and rewarded instead of guessing, and platforms could reward their high quality content creators more?

TikTok’s challenge of fair value sharing motivated us to build Alto: a system that always puts creators and high content quality first, combining modern AI-powered video analysis with a fair and transparent content scoring system in a modern dashboard.

What it does

Alto is a creator-centric value sharing platform that rewards quality and authenticity as much as popularity. Unlike systems that reward only virality or raw view counts, Alto introduces a transparent and explainable ecosystem where creators know exactly why and how they are rewarded. Alto does even more: it offers content creators tips on how to improve their content quality, warns them of possible compliance issues, and logs every transaction in a secure ledger tied to the creator.

Key Features:

  1. Creator Dashboard (Frontend – Lynx.js)

    • Splash & Landing: Animated coin splash, platform stats (e.g., active creators, total payouts).
    • Creator Registration: Sign up with display name, email, handle, and niche. Smooth onboarding into the dashboard.
    • Video Management:
      • Add TikTok videos by link.
      • View Quality Scoring Engine (QSE) breakdown (Retention, Engagement, Niche Fit, Authenticity).
      • Explore preloaded demo videos with instant analysis.
    • Video Analysis: Detailed scoring, compliance checks, and improvement suggestions (e.g., “great hook, weak originality”).
    • Wallet Management: Track credits → Nanas earned, with 24h/7d charts.
    • Withdrawals: Redeem Nanas into USD via PayPal, bank transfer, or gift cards. ($5 min; >$50 requires KYC).
    • Analytics & Insights: Interactive charts showing earnings, QSE evolution, and performance trends.
    • Profile System: Creator stats, tiering, and account settings.
    • Cross-Platform: Single codebase for iOS, Android, Web; responsive design and near-native performance.
  2. In-App Currency: Nanas 🍌

    • Two-Step Redemption: Viewer actions → credits → Nanas. Creators later cash out. This prevents fraud and money laundering.
    • Dynamic Value (future): High-quality creators earn Nanas with higher purchasing power.
    • Auditability: All transfers logged in a transparent ledger for regulator-friendly oversight.
  3. Backend Pipeline (Python + Gemini + QSE)

    • Video Evaluation Agent (Google Gemini + heuristics):
      • Hook, Retention, Clarity, Originality, Niche Fit, Engagement.
    • Compliance Agent:
      • Flags age-inappropriate content, hate speech, harassment, violence, misinformation, scams/fraud, IP/privacy issues.
    • Structured JSON Reports: Includes synopsis, metric scores, compliance flags, payout tier, and recommendation.
    • Batch Processing: Process multiple .mp4 videos at once with scripts; outputs saved to JSON with terminal summaries.
    • Customizable Weights: Adjust payout/score parameters in utils.py.
  4. Explainability & Transparency

    • Every score comes with reasons (e.g., “strong retention,” “weak originality,” “possible misinformation”).
    • Ensures creators understand what drives their success and what to improve.
  5. Demo-First Experience

    • Preloaded sample videos and mock viewer events allow immediate testing.
    • Judges can see payouts, compliance checks, and dashboards instantly without setup.

Value Flow: Viewer actions (watch, like, comment, share) + AI/QSE assessment → Score → Credits minted to creator’s Alto wallet → Credits converted to Nanas → Creators cash out or spend Nanas in-app.

How we built it

  • Backend (Python + Gemini): Video processing pipeline for .mp4 files or TikTok links. Outputs JSON reports with scores, flags, and payout recommendations.
  • Frontend (Lynx): Interactive dashboard showing video previews, credit balances, and compliance status.
  • Algorithms: Combination of Gemini model outputs with our in-house Quality Scoring Engine (QSE), blending metrics like engagement depth, niche fit, and authenticity.
  • Data & Demo: Sample videos and mock viewer events for instant testing; supports batch processing of uploaded content.

Challenges we ran into

  • Time crunch: Connecting AI evaluation, heuristic scoring, and the dashboard in under 72 hours.
  • Integration trade-offs: Skipped deep TikTok integration in favor of Gemini evaluation + previews for reliability.
  • Balancing AI vs heuristics: Ensuring Gemini insights and QSE scoring reinforced each other instead of clashing.

Accomplishments that we're proud of

  • End-to-End Prototype in 72 Hours
    Built a fully working pipeline, from video evaluation to payout dashboards under intense time pressure, connecting frontend, backend, and AI scoring seamlessly.

  • Quality Scoring Engine (QSE)
    Designed and deployed a transparent scoring system that doesn’t just assign numbers but explains why content earns rewards, balancing AI insights with human-readable heuristics.

  • Creator-Friendly Dashboard
    Delivered a sleek, responsive Lynx.js dashboard where creators can register, upload videos, track scores, view compliance checks, and cash out Nanas.

  • Fraud-Safe Currency System
    Implemented the two-step credit → Nanas model, preventing abuse while keeping cashouts smooth and regulator-friendly.

  • Transparency & Explainability
    Every payout and compliance flag is backed by clear reasoning, making Alto not just fair, but also trustworthy to creators, platforms, and regulators.

  • Demo-Ready Experience
    Packed Alto with preloaded videos, instant scoring, and mock payouts so judges and testers can experience the full value-sharing cycle without extra setup.

What we learned

We learned how hard it is to balance content quality, authentic engagement, and compliance in a single pipeline.

  • Experimented with both heuristic scoring (QSE) and Gemini for nuanced evaluation.
  • Realized explainability is essential: numbers alone aren’t enough without reasons (“great hook,” “weak originality,” “possible misinformation”).
  • Saw how compliance checks (safety, hate speech, misinformation) need to integrate seamlessly with engagement metrics.

What's next for Alto

With our hackathon project, we aim to push one step further and continue developing and refining our platform, both the scoring algorithm as well as the improving the creator experience. In the future, we have ambitions to partner Alto as a service alongside global content platforms that are interested in fostering a better value sharing ecosystem, both for short and long form content.

Built With

+ 17 more
Share this project:

Updates