Phase 10Days 113-117
Image / File Storage at Scale
Cloudinary, S3 + CloudFront, signed URLs, direct-to-storage uploads
Phase Goal
Move beyond "save a file path in Mongo" — handle uploads, transformations, signed URLs, and CDN delivery the way real apps do.
Progress
Notes
Open the written lectures for this course before checking off the phase topics.
Projects
Oral-History Archive Ingest — DEPLOYED
Day 113: The Storage Landscape
Day 114: Cloudinary Done Right
Day 115: AWS S3 + CloudFront
Day 116: Image Pipeline & Optimization
Day 117: Project — Oral-History Archive Ingest
Capstone
Oral-History Archive Ingest — DEPLOYED
Build archive intake for interview audio, consent forms, portraits, and transcripts: direct browser uploads, private masters, signed cataloguer links, public-safe derivatives, checksum metadata, and withdrawal rules. Use S3-compatible storage, presigned URLs, appropriate CDN delivery, and document the threat model.
- Direct-to-storage upload with server-issued signed policy/URL and MIME/size validation.
- Private originals, expiring reviewer links, and public-safe thumbnail derivatives.
- Upload progress, failure recovery, and a concise storage-security README.
Phase Complete!
After this phase, you'll be able to:
- Choose Cloudinary vs S3+CloudFront consciously
- Implement direct-to-storage uploads with signed URLs
- CloudFront fronting S3 with OAC
- Image pipeline: resize, format, blur placeholder
Your apps can handle uploads at production scale without melting the server.