Lo scopo principale è creare un generatore di “Stories2 verticali a partire da contenuti long-form (talk, sport, news).

La pipeline comprende: trascrizione, segmentazione in scene, riassunto e copy/CTA con IA generativa, montaggio automatico con template grafici (titoli, sottotitoli, bumper) e export in 9:16/1:1.

Previste guardrail di brand-safety e un piccolo cruscotto per l’editor human-in-the-loop.

Argomento principale: Data Science, Scene detection, Summarization, Rendering (FFmpeg/Remotion).

Corso di studio e requisiti candidati: Informatica, Ingegneria Informatica. JS/Python, basi di video processing; gradite competenze React.js.

Sede tirocinio: Napoli e Milano.

__________

The main goal is to create a generator of vertical "Stories" from long-form content (talks, sports, news).





The pipeline includes: transcription, scene segmentation, summarisation and copy/CTA generation with Generative AI, automatic editing with graphic templates (titles, subtitles, bumpers), and export to 9:16/1:1.

Brand-safety guardrails and a small dashboard for the human-in-the-loop editor are expected.

Main Topic: Data Science, Scene detection, Summarisation, Rendering (FFmpeg/Remotion).

Course of Study and Candidate Requirements: Computer Science or Computer Engineering. JS/Python, basics of video processing; React.js skills are a plus.

Internship Location: Naples and Milan.

📌 Auto-Stories for Social: From Long-Form to 9:16 (Italia)
🏢 Kineton
📍 Italia