A comprehensive developer relations strategy and interactive presentation for CrewAI - the open-source framework for orchestrating autonomous AI agent crews.
π Live Site: [Coming Soon on Vercel]
π Presentation: how-it-works.html
π€ CrewAI: crewai.com | GitHub
This project showcases a world-class developer relations strategy for CrewAI, featuring:
- Comprehensive DevRel Strategy: Objectives, pillars, programs, metrics, and 90-day execution plan
- Interactive Presentation: Slide deck with keyboard navigation, speaker notes, and Web Speech API narration
- Real-World Use Cases: Multi-agent patterns with code examples and architecture diagrams
- Ecosystem Guide: LLM providers, tools, memory systems, and deployment patterns
- Production-Ready: Static site, self-hosted assets, accessible, and performant
# Clone the repository
git clone https://github.com/colygon/crewai-devrel.git
cd crewai-devrel
# Open in browser
open index.html
# Or serve with Python
python3 -m http.server 8000
# Visit http://localhost:8000# Install Vercel CLI
npm i -g vercel
# Login and deploy
vercel login
vercel --prodcrewai-devrel/
βββ index.html # Main DevRel strategy page
βββ how-it-works.html # Interactive presentation
βββ use-cases.html # Multi-agent patterns
βββ ecosystem.html # Integrations and tools
βββ assets/
β βββ css/
β β βββ styles.css # Main stylesheet
β β βββ presentation.css # Presentation-specific styles
β βββ js/
β β βββ main.js # Site behaviors
β β βββ presentation.js # Slide deck engine
β β βββ diagrams.js # Architecture diagrams
β βββ vendor/
β βββ highlightjs/ # Syntax highlighting (local)
βββ data/
β βββ metrics.json # DevRel KPIs and metrics
βββ vercel.json # Deployment configuration
βββ README.md # This file
- Five Pillars: Content, Code, Community, Enablement, and Feedback
- Programs: Weekly recipes, biweekly livestreams, monthly hackathons
- 90-Day Plan: Foundation β Amplification β Acceleration
- Metrics Dashboard: Real-time KPIs from data/metrics.json
Keyboard Controls:
βorSpace- Next slideβorShift+Space- Previous slideF- Toggle fullscreenN- Toggle speaker notesEsc- Exit fullscreenHome/End- First/last slide
Features:
- Progress bar and slide counter
- URL hash bookmarking (
#slide-5) - LocalStorage progress saving
- Web Speech API narration (toggle on/off)
- Reduced motion support
- Semantic HTML5 landmarks
- ARIA labels and roles
- Keyboard-only navigation
- Color contrast WCAG AA compliant
- Reduced motion support
- Focus indicators
Edit use-cases.html and add a new section:
<div class="use-case-card">
<h3>Your Use Case Title</h3>
<p>Problem statement and outcome</p>
<!-- Architecture diagram -->
<div class="diagram-container">
<!-- Agent and task nodes -->
</div>
<!-- Code example -->
<pre><code class="language-python">
# Your CrewAI code here
</code></pre>
</div>Edit how-it-works.html and add:
<section class="slide" data-slide-id="N">
<div class="slide-content">
<h2>Your Slide Title</h2>
<p>Content...</p>
</div>
<div class="slide-notes" style="display: none;">
Speaker notes for narration
</div>
</section>Update totalSlides in the presentation controller.
Edit data/metrics.json with current stats:
{
"github": {
"stars": 25000,
"forks": 3200,
...
},
...
}The site automatically loads and displays these metrics.
Colors:
- Primary:
#06D6A0(Turquoise) - Accent:
#7C3AED(Purple) - Cyan:
#22D3EE - Background:
#0B1020β#0F1629
Typography:
- Base: System fonts (SF Pro, Segoe UI, Roboto)
- Mono: SF Mono, Monaco, Cascadia Code
Spacing Scale:
--space-xsto--space-3xl- Based on 0.25rem increments
- No external CDNs: All assets self-hosted
- Minimal JavaScript: <10KB total
- Optimized CSS: Scoped, efficient selectors
- Fast load times: <1s on 3G
Target Lighthouse scores:
- Performance: 95+
- Accessibility: 95+
- Best Practices: 100
- SEO: 100
Contributions welcome! Please see CONTRIBUTING.md for guidelines.
- Fork the repository
- Create a feature branch (
git checkout -b feature/amazing-addition) - Make your changes
- Test locally
- Commit (
git commit -m 'feat: add amazing feature') - Push (
git push origin feature/amazing-addition) - Open a Pull Request
MIT License - see LICENSE for details.
- CrewAI Team: For building an amazing multi-agent framework
- Perplexity DevRel: Inspiration for presentation structure
- Community: For feedback and contributions
Colin Lowenberg
AI Developer Relations & Platform Ecosystems
- LinkedIn: colinlowenberg
- GitHub: @colygon
- Twitter: @colinlowenberg
Built with β€οΈ for the CrewAI community
This project demonstrates a comprehensive approach to developer relations for AI frameworks, combining strategy, education, and community building.