Season 6: Build Code Nexus Assist · Episode 3 of 8
Creators Want Visual Superpowers
You play: Junior AI Developer
Included with Core programme and Certification prep
The situation
Two asks reached Jax this week. The first is accessibility: Naledi, a ceramics creator, has followers who use screen readers, and her feed reads to them as "image, image, image". The second is convenience: creators want to brighten a product shot or soften a busy background without opening a design tool. Code Nexus Assist can now look at pictures and video — describe them, and draft edits and new images. The team has to decide what it may say about a picture, what it may change in one, and who checks the results.
What you'll learn
- AI-103 vision and media: reading image and video analysis (text, objects, regions, scene, confidence), writing supported alt text and visual descriptions, video understanding as timed segments, masked edits limited to a region, image generation controls and policy, embedded-text injection, provenance labelling and human review.
- Inpainting and mask-based edits, video generation and editing, Content Understanding single-task and pro modes, object and region detection, and indirect prompt injection through images.
Who you work with
Jax Rivera
Creator Partnerships Lead
Maya Chen
Senior AI Engineer · Your mentor
Lydia Roe
Security Engineer
Scenes
- 1.Two asks, one rule
- 2.Say only what the picture shows
- 3.The description batch
- 4.Edit the background, not the truth
- 5.The banner that talks back
- 6.Vision check
What you leave with
You kept the fictional photo description within the supplied evidence, implemented the batch formatting and review rules, and applied the simulated mask and provenance controls. The injected banner goes to review and the fabricated endorsement is blocked. Real detections, masks and labels still need source checks and delivery validation.