Snehansu Panda
As organizations rush to integrate large language models (LLMs) and autonomous AI agents into their data ecosystems, many encounter a common pitfall: expecting AI to "just work" the moment it is connected to a database. However, AI agents lack inherent organizational context and are constrained by strict processing limitations, such as context windows. To truly unlock the productivity and accuracy of AI applications, organizations cannot rely on raw data alone; they must bridge the gap between their data products and AI comprehension by engineering a highly optimized, AI-ready semantic layer.
Join this session with Snehansu Panda, senior data engineer at Denodo, to explore the architecture and principles behind building this crucial semantic bridge. We will dive into how modern data management platforms empower teams to curate high-quality metadata, control data exposure, and integrate data seamlessly.
Attend & Learn:
- Guidelines for selecting, normalizing, and exposing business views relevant to specific AI use cases
- Best practices for view and field naming to eliminate ambiguity and reduce AI processing overhead
- An actionable framework for writing concise, effective view and field descriptions that define the scope and use cases of your data
