{
  "video_id": "Obaq2ddTYRs",
  "title": "How Cemex Builds with LlamaIndex to Transform Operations, Supply Chain, and Customer Experience",
  "url": "https://www.youtube.com/watch?v=Obaq2ddTYRs",
  "transcript": "Since we started using Lama Cloud, our data ingestion process moved from taking around three weeks up to less than a day. My name is Daniel Garcia Zapata. I'm a senior data scientist at Cemex, a global building material company, one of the largest red mix producers in the world. We use AI to improve our maintenance, supply chain optimization, smart operations, health and safety and for commercial efforts we use it to help our salespeople improve their engagement with our customers. We didn't have a standardized way to build agentic solutions. We had an in-house solution that took weeks just to incorporate new data into agents. We used Lama Index to build tools and agents that integrate seamlessly with our internal systems. We chose Lama Index because it directly addresses some of the key payments that we have found in other frameworks. First, with document ingestion. We built an in-house solution that didn't scale well and was hard to maintain. Working with high technical dense documents was a challenge. Since we started incorporating LamaPars, we were able to extract and structure the data and we realized that it dramatically improved the solution and response of our agents. We also observed a 20% response in shock relevance and accuracy without making any changes in our downstream agents. Second, it simply integrates well with our current stack, such as Databricks, MFLOW, and MongoDB. Finally, the level of abstraction and customization that Lama Index provides has helped us to extend with in-house modules without disrupting the development process. For us, Lama Index fits perfectly with how we build and envision AI solutions in our organization. Thank you.",
  "transcript_chars": 1686,
  "transcript_filled_at": "2026-06-06T14:35:51.038046+00:00",
  "transcript_filled_by": "tk-bulk-groq-retry-20260606"
}