~/talk

Editorial AI in AEM: Giving Editors Control Over LLM-Generated Image Descriptions

Talkat adaptTo() 2026 · 2026-09-28..09-30· [ai-assisted-development]
links:none on file
slides:none on file

An open-source AEM workflow encodes images in Base64 and sends them over plain HTTP to Azure AI Foundry for vision-capable LLM chat completions, returning alt text saved into asset dc:description metadata; prompts live as plain-text files inside the DAM so authors get versioning, permissions, diffs and immediate edits. Implementation adds caching, circuit breakers and timeouts, costs ~5¢ per asset (2,000 assets ≈ $100) and replaces ~167 manual hours; prototype ships as a Maven module and took about a week of dev time.

AI-written summary · about

~/search
type to search 6,518 conferences, 143,918 speakers and 192,133 talks
~/help
cmdctrl k
search every conference, speaker and series
/
focus the filter search
h
home
c
conferences
s
speakers
e
series
t
light / dark — the status bar button does the same
← →
on a talk: 5 seconds back / forward
0–9
on a talk: jump to 0%–90% of the recording
?
this help
esc
close this, then clear every filter — or just the search, while you are typing in it

Ignored while you are typing. Pages without a filter rail have no search to focus.