4️⃣ Week 4 — AI Services & Capstone 🏆
4️⃣ Week 4 — AI Services & Capstone 🏆
Theme: Build AI into your apps. Then ship your capstone portfolio piece. Live session: Sat 13 Jun 2026 · 3 PM GMT Time commitment: 2 hr live + 6-8 hr labs (capstone is bigger than usual)
🎯 By the end of this week, you’ll have
- ✅ A Smart Document Analyzer that processes any PDF with Textract + Comprehend
- ✅ A complete AI Customer Feedback Hub — your portfolio piece
- ✅ Hands-on experience with 4+ AWS AI services
- ✅ A LinkedIn-ready project to anchor your CV around
📚 This week’s content
| Item | Type | Time | Link |
|---|---|---|---|
| 🎬 Week 4 Slides | Live session | 2 hr | slides/ |
| 🛠️ Lab 7 — Smart Doc Analyzer | Guided lab | 90 min | lab-7-doc-analyzer/ |
| 🏆 Lab 8 — AI Capstone | Build-on-your-own | 3-5 hours | lab-8-ai-capstone/ |
| 🎯 Week 4 Quiz | Interactive (FINAL!) | 15 min | quiz/ |
📅 Suggested schedule
| Day | What to do |
|---|---|
| Sat 13 Jun | Final live session 3 PM GMT |
| Sun 14 Jun | Lab 7 (Smart Doc Analyzer) |
| Mon-Thu | Lab 8 (The CAPSTONE — give it the time it deserves) |
| Fri 19 Jun | Take the final quiz + post your capstone to LinkedIn |
| Sat 20 Jun | Demo day for group projects 🏆 |
🧠 Concepts we’ll cover
- AWS AI service overview: Textract, Comprehend, Rekognition, Polly, Translate, Transcribe
- Textract — extracting text + tables + forms from any document
- Comprehend — sentiment, entities, key phrases, language detection
- Rekognition — object detection, face analysis, content moderation
- AI service chaining — combining multiple services to solve real problems
- API Gateway — HTTP endpoints for Lambda functions
- DynamoDB — serverless NoSQL for flexible JSON data
💼 The Capstone — why it matters
Lab 8 isn’t “another lab”. It’s the project that goes on:
- Your LinkedIn (under Featured / Projects)
- Your CV (under Personal Projects)
- Your GitHub (pinned)
It’s the difference between saying “I learned AWS” and “Here’s a live AI app I built and shipped.”
The capstone takes 3-5 hours. Don’t rush it. Polish > speed.
🆓 Free tier note
| Service | Free amount |
|---|---|
| Textract | 1,000 pages/month for 3 months |
| Comprehend | 50K text units/month (12 months) |
| Rekognition | 5K image API calls/month (12 months) |
You won’t blow through these unless you process huge volumes.
📦 Folder contents
week-04-ai-capstone/
├── slides/
│ └── Week4_CloudWithShad.pptx
├── lab-7-doc-analyzer/
│ ├── README.md
│ ├── Week4_Lab7_Smart_Doc_Analyzer.html
│ └── starter-code/
│ └── analyze_document.py
├── lab-8-ai-capstone/
│ ├── README.md
│ ├── Week4_Lab8_AI_Capstone.html
│ └── starter-code/
│ ├── frontend/
│ │ └── index.html
│ ├── lambda/
│ │ ├── process_feedback.py
│ │ └── dashboard_stats.py
│ └── dashboard/
│ └── dashboard.html
└── quiz/
└── Week4_Quiz_AI_Capstone.html
🎓 Capstone submission requirements
For your capstone to count:
- Live URL of your form (S3 hosted)
- Live URL of your dashboard (S3 hosted)
- GitHub repo with all your Lambda code + frontend HTML
- README in the repo explaining the architecture + how to run it locally
- Screenshot of the dashboard showing real data after at least 5 submissions
- LinkedIn post featuring the project — tag #cloudwithshad
Drop everything above in the WhatsApp group with #capstone-done.
🏆 What’s next: Group projects + Demo day
You’re not quite done yet. Mon 15 Jun: group project assignments drop. You’ll have one week to build a polished Streamlit app on EC2 with your team. Demo day is Sat 20 Jun 2026 · 3 PM GMT — bring your A-game.
→ See group-projects/ for the three project options + how to pick a team.
⬅️ Previous: Week 3 — Scaling & DevOps 🏆 Next: Group Projects ⬅️ Back to repo