1 day · From LLMs to autonomous agents · Designed for business leaders & cross-functional teams
A single-day deep dive into Agentic AI — from foundational concepts to building production-ready agent systems on AWS.
What is agentic AI, how it differs from chatbots, workflow patterns, and insurance use cases
4 workflow patterns, AWS tools landscape, how AI connects to your systems (MCP), live demos
Agent Design Canvas — design an AI agent for your own workflow (no coding required)
Portfolio Managers, Product Owners — identify where AI agents can accelerate your workflows
Audit executives, VP Security, QA teams — understand AI guardrails, oversight, and governance
Head of Cloud COE, DevOps, Head of Platforms & Digital UX — evaluate tools and implementation paths
HR leaders, Distribution Program Management, Management Associates — explore AI for operational efficiency
The centerpiece of this workshop — design an AI agent for a real workflow from your own team. No coding required.
Pick a workflow from your team → choose a pattern → fill out the Agent Design Canvas → submit for AI scoring on the leaderboard
The workshop follows the AWS Agentic AI Foundation curriculum, customized with AnyCompany Insurance examples throughout.
| Module | Topic | Key Concepts |
|---|---|---|
| M0 | Introduction | Workshop goals, environment setup, agenda |
| M1 | From LLMs to Agents | LLM limitations, perception-reasoning-action, fallacy of composition |
| M2 | Exploring Agentic AI | Agent types, autonomy levels, memory, tool use, reflection |
| M3 | Agentic Workflows | Chaining, parallelization, routing, orchestration — insurance examples |
| M4 | AWS Developer Tools | Bedrock Agents, AgentCore, Kiro IDE, MCP integration |
| M5 | Frameworks | Strands Agents SDK, LangGraph, CrewAI — when to use which |
| M6 | Custom Solutions | Production patterns, guardrails, evaluation, human-in-the-loop |
| M7 | Wrap Up | Key takeaways, next steps, resources |
Visual, interactive walkthroughs of key concepts. Use these as reference during and after the workshop.
Understanding the internals helps you design better agents. These cover the three core stages of how AI processes text.
Agentic AI use cases explored in this workshop:
| Domain | Use Case | Pattern |
|---|---|---|
| Claims | FNOL intake & triage | Routing + Tool Use |
| Claims | Medical claims auto-adjudication | Parallelization + Orchestration |
| Claims | Fraud detection & investigation | Chaining + Tool Use |
| Underwriting | Automated life insurance underwriting | Chaining + Parallelization |
| Underwriting | Group insurance scheme quoting | Chaining + Tool Use |
| Policy | Intelligent policy change agent | Routing + Tool Use |
| Policy | Lapse prevention & reinstatement | Chaining + Tool Use |
| Compliance | MAS regulatory reporting | Chaining + Parallelization |
| Compliance | AML/CFT screening & SAR drafting | Chaining + Tool Use |
| Actuarial | IFRS 17 reporting automation | Chaining + Parallelization |