Glossary

Insights, News & Updates
Sandboxed Agent Execution refers to the practice of running an agentic AI system within a restricted, isolated environment that limits its access to external systems, data, tools, and resources. 
A savings plan is a cost-reduction pricing model offered by major cloud services providers.
Savings rate in cloud computing refers to the percentage reduction in costs achieved by optimizing resource.
Secure Multi-Party Computation (SMPC) is a cryptographic method that allows multiple parties to jointly compute a function over their private inputs without revealing those inputs to one another.
Self-attention is a technique used in neural networks, especially in transformer models.
Self-consistency decoding is a method used in natural language generation.
Self-supervised learning is a type of machine learning.
Semantic Memory in agentic AI refers to the persistent memory system that enables autonomous agents to store, organize, retrieve, and reason about generalized knowledge, including facts, concepts, rules, relationships, and domain understanding, independent of specific experiences or episodes.