Taylor Riggan

AI agents assert relationships such as compatible products, cleared counterparties, or approved suppliers, derived from LLM reasoning rather than verified data. When these assertions are wrong, downstream business decisions fail.

Most knowledge graphs answer questions but never close the loop with operational systems. In this session, explore how to build an ontology-driven context layer that transforms a knowledge graph into an active control plane for AI agents.

Large language models pattern-match on tokens, not business meaning, which causes subtle failures when applied to enterprise data. Ontologies address this by giving AI systems a formal structure for knowledge: a graph defining what exists, how entities relate, and what questions are meaningful.

Retrieval Augmented Generation (RAG) applications use the power of generative AI to analyze private datasets. GraphRAG combines knowledge graphs with RAG to produce explainable responses that are grounded in the semantic relationships between concepts, entities, and the underlying content.

The emergence of generative AI has also seen the rise of innovative new graph workloads like GraphRAG and natural language querying. In this chalk talk, learn how to deploy these new workload types on Amazon Neptune.

In this chalk talk, demystify the process of migrating your self-managed InfluxDB open source database deployment to Amazon Timestream for InfluxDB. Whether you're currently running InfluxDB versions 1.X or 2.X, discover the various options and steps to make the transition seamless.

Retrieval Augmented Generation (RAG) applications use the power of generative AI to analyze private, differentiating datasets. However, baseline RAG can sometimes produce poor responses that lack explainability and contextual awareness and include conflated sources and spurious claims.

Use cases such as real-time recommendation engines demand high performance. When performance is your top priority, model optimizations can be very important to achieving your goals.

In this workshop, learn to build modern web applications at scale using purpose-built databases. Discover how to apply development patterns using Amazon DynamoDB, Amazon ElastiCache, Amazon Neptune, and Amazon OpenSearch Service to build a fully functional and scalable bookstore ecommerce application and dive deep into best practices along the way.

Amazon Neptune is a fast, reliable, and fully managed graph database service that makes it easy to build and run applications that work with highly connected datasets. In this workshop, get hands-on experience and learn how to use Amazon Neptune to create and load a connected graph dataset, write connected graph queries with Gremlin and openCypher, and generate rich visualizations with your...

Graph databases have many uses including ways to detect fraud, discover security risks, and store data lineage across disparate datasets. Building such applications can require an iterative approach, and many developers prefer to start this process on their local workstations.

Join us for a practical, hands-on workshop on using Amazon ElastiCache. This session is designed for developers, engineers, and database administrators who are involved in designing ElastiCache for Redis.

In this chalk talk, learn how to build a cloud-centric, graph-based identity resolution system that connects customer data across devices, channels, and touchpoints and helps enable better media buying and personalization. Learn about best practices and the mistakes to avoid for identity resolution data collection, processing, and preparation.

In this session, get hands-on with Amazon Neptune and build a cloud-based graph application. Learn how to quickly load data and begin writing Gremlin traversals.