Teradata (TDC) Status Update summary

Business and technical background

• AI-driven tools are transforming business value, but require robust analytic and experimental foundations for success.

• Key pillars for AI adoption include flexibility, productivity, cost-effectiveness, and scalability.

• Challenges include IT legacy constraints, budget limitations, data access concerns, and operationalizing experimental models.

• Data engineers, scientists, and developers face evolving requirements for handling large, complex datasets and integrating AI into applications.

Teradata AI Unlimited solution overview

• AI Unlimited offers on-demand, cloud-native AI and ML capabilities, eliminating infrastructure management and cost overruns.

• Supports rapid experimentation and seamless transition from experiment to production using the same analytic engine.

• Enables access to data across clouds and open table formats, supporting broad enterprise data needs.

• Fully integrates with enterprise GitHub/GitLab for state maintenance and authentication.

Demonstration and experimental workflow

• Demonstration focused on using generative AI and open-source tools to extract business value from retail customer reviews.

• Customer reviews are vectorized using a 50-dimensional GloVe model and processed with native Word Embeddings.

• K-means clustering segments customers based on semantic review patterns, offering new segmentation insights.

• Experiments are run iteratively and in parallel, leveraging the platform's scalability.