Skip to content

Conversation

@Vajbratya
Copy link

Summary

This PR introduces a complete launch video for Laudos.AI built with Remotion and Next.js. The video showcases the product's key features through a 5-scene narrative that highlights the problem radiologists face, the solution's capabilities, and a live demo of the AI-powered laudo generation system.

Key Changes

  • Project Setup: Added Next.js 16 configuration with Remotion integration, including webpack overrides and transpilation of shared packages

  • Video Composition: Created a full 45-second launch video (FullVideo.tsx) composed of 5 sequential scenes:

    • Intro Scene: Problem statement with animated word-by-word text revealing ("Radiologists spend hours writing reports")
    • Problem Scene: Visual representation of legacy software challenges and lack of standardization
    • Features Scene: 2x2 grid showcasing 4 core features (LaudAI, Agentic AI, Voice Dictation, Crit Enterprise)
    • Demo Scene: Interactive mockup showing voice-to-structured-report workflow with critical finding alerts
    • Outro Scene: Call-to-action and closing message
  • Design System: Implemented clean, professional white design with:

    • Consistent color palette (blue accent #0066FF, red for critical alerts)
    • Smooth animations using Remotion's interpolation and easing functions
    • Responsive typography with Inter and JetBrains Mono fonts
  • Audio: Integrated 9 sound effects (whoosh, pop, typing, etc.) synchronized with animations for enhanced engagement

  • Web Preview: Added Next.js page (/src/app/page.tsx) with Remotion Player for previewing the video in development

  • Styling: Configured Tailwind CSS v4 with PostCSS for the Next.js app

Notable Implementation Details

  • All animations use frame-based timing for precise synchronization with audio
  • Staggered animations create visual hierarchy (e.g., feature cards appear sequentially)
  • Critical alert component demonstrates enterprise feature with pulsing animation
  • Voice dictation demo includes animated waveform visualization
  • Sound effects are conditionally rendered using Remotion's Sequence component
  • Shared packages (@launchpad/shared, @launchpad/assets) are transpiled for compatibility

Scripts

  • pnpm dev: Run Next.js development server with video preview
  • pnpm remotion: Open Remotion Studio for full editing
  • pnpm render: Render final video at full quality
  • pnpm render:preview: Render at 50% scale for quick testing

Create a professional launch video for laudos.ai featuring:
- IntroScene: Problem statement about radiologists spending hours on reports
- FeaturesScene: Showcase unique features (LaudAI, Agentic AI, Crit Enterprise)
- DemoScene: Product UI mockup with voice dictation and structured reports
- OutroScene: CTA with company branding

Design follows clean white background with Inter font for a professional look.
Highlights key differentiators: LaudAI (proprietary AI), multi-agent architecture,
and Crit for critical findings detection in enterprise settings.
New scene highlights problems with legacy radiology software:
- Mockup of outdated Windows 2000-era interface
- List of problems: ultrapassadas interfaces, laudos sem padronização,
  100% trabalho manual, sem inteligência artificial
- Visual contrast sets up Laudos.AI as the modern solution

Video now flows: Intro -> Problem (competitors) -> Features -> Demo -> Outro
- IntroScene: "Radiologistas gastam horas escrevendo laudos. Todos. Os. Dias."
- IntroScene: "E se a IA pudesse ajudar?"
- OutroScene: Stats em português (Nuvem, Compliance, Potência)
- Demais cenas já estavam em PT-BR
- Corrigir "agentica" para "agêntica" em todas as cenas
- DemoScene: nova interface completa de editor de laudos
  - Header com logo, info do paciente, status LaudAI
  - Sidebar esquerda com informações do exame
  - Editor central com seções TÉCNICA, ACHADOS, IMPRESSÃO
  - Toolbar de formatação com botão de ditado
  - Sidebar direita com sugestões da LaudAI (formatação, validação, estilo)
  - Botões de ação: salvar rascunho e assinar laudo
Atualiza DemoScene para corresponder exatamente à interface real do
copilot.laudos.ai:
- Sidebar esquerda com "Laudos Recentes"
- Barra de busca com atalho ⌘K
- Botão de microfone "Ditar laudo" com animação
- Seletor de templates e modelos frequentes
- Toolbar com formatação (B, I, U, H1, H2)
- Toolbar secundário com Buscar Artigos, Extrair Frases, Crit
- Barra de atalhos no rodapé
- Animação de digitação de laudo
Fluxo correto:
1. Usuário clica no botão "Ditar laudo"
2. Mic fica vermelho (gravando)
3. Ondas de voz aparecem durante gravação
4. IA transcreve em tempo real enquanto usuário fala
5. Laudo estruturado: TÉCNICA, ACHADOS, IMPRESSÃO
6. Botão salvar fica verde ao finalizar
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants