Document Processing · Marketing Personalization
Astro needed faster and more consistent metadata generation across its web and mobile platforms, but manual tagging slowed content operations, limited scale, created classification inconsistencies, and reduced the quality of the personalization signals used across its OTT experience.
Reported impact
Astro automated a labor-intensive metadata function, dramatically cut manual tagging work, improved consistency and throughput across content operations, and strengthened the personalization signals that support a smoother digital entertainment experience for viewers.
Astro partnered with TCS to turn metadata tagging into an agentic AI workflow instead of a manual media-operations task. TCS used TCS AI VX Studio as the orchestration layer, running on Amazon Bedrock with the NOVA multimodal model so autonomous AI agents could inspect web and app visuals, detect interface elements, and generate semantic metadata such as titles, genres, and categories. Product and content teams stay in the loop through a human-feedback process that corrects outputs and continuously improves prompt behavior as layouts and content formats change. The modular architecture processes assets in parallel for scale, then feeds the enriched metadata into Astro's downstream analytics and recommendation workflows, giving the broadcaster a repeatable way to tag thousands of assets while keeping the model adaptive to new UI patterns.
Visual content from Astro's web and mobile platforms is processed by autonomous AI agents running through TCS AI VX Studio on Amazon Bedrock with the NOVA multimodal model. The agents detect UI elements and generate semantic metadata, product and content teams feed corrections back through the human-in-the-loop process, and the enriched tags flow into downstream analytics and recommendation systems.
"90% reduction in content-metadata management effort"
"AI agents analyse web and app visuals, detect UI elements, and generate semantic metadata"
"Leveraging TCS AI VX Studio, Amazon Bedrock, and the NOVA multimodal model, TCS implemented an agentic AI-powered system"
"a projected two-point rise in net promoter score (NPS) and a 20% improvement in customer satisfaction"
"fivefold increase in content classification"
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
Related
Google Cloud Sample Archived · Sep 25, 2026 TapNation However, the company quickly realized that the true revolution lay elsewhere. "For us, AI isn't just another tool stacked on top of existing ones. It is the backbone of a new operating model, capable of supporting all… 60% reduction in time spent drafting technical specifications Gemini · Gemini Enterprise · Gemini Enterprise Agent Platform Google Cloud Sample Archived · Sep 9, 2026 The Points Guy As teams shift from generative AI experimentation to production, scaling video creative assets has emerged as a major operational bottleneck. While ad platforms allow for strategic audience targeting, the production of… 71,000+ clicks and 67,000+ conversions driven by GenAI video assets Gemini · Veo 
OpenAI Sample Aug 20, 2025 MIXI MIXI needed a secure company-wide AI environment so employees could share GPTs, use internal data safely, and speed up creative, investment, and communication work across a diverse consumer-services portfolio. 45-day deployment ChatGPT Enterprise · Copy Checker GPT · Creative Planning GPT 
