From the course: Building Agentic AI Systems
Join today to access over 24,900 courses taught by industry experts.
Putting together the reference architecture
From the course: Building Agentic AI Systems
Putting together the reference architecture
- Let's create a reference architecture using AWS services for the agent AI system we are building for our scenario. At the foundation lies the data layer, where we use Amazon S3 and AWS Glue to bring in structured data from internal systems, like HRIS and LMS, along with unstructured data, such as peer feedback and discussion logs. Real-time data streams from user interactions, click patterns, and assessments are ingested through Amazon Kinesis. External APIs connect to sources like LinkedIn Skills Insights and LinkedIn Learning to enrich the dataset. The processing layer brings intelligence into the system. It uses large language models like OpenAI and LLaMA through AWS Bedrock for analyzing skill gaps, personalizing learning, and understanding feedback. For deeper reasoning models like SOAR and BDI help agents plan and act. AWS Step Functions and EventBridge coordinate agent workflows. For example, they activate the personalization agent after the skill gap agent detects the gap…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.