Implementation Week 2 · Saturday

Day 13: Hands-On AgentCore + Service Comparison Drilling

Learning Objectives

0/5 completed

Blog30m
Deploy Production AI Agents in 2 Commands
AgentCore CLI: agentcore create, agentcore deploy, test endpoint.
Blog25m
AgentCore Memory Management (Mike Chambers)
Short-term vs long-term vs episodic memory. Code walkthrough.
Blog20m
AgentCore Code Interpreter
Secure sandbox for Python execution. Supports pandas, numpy, matplotlib.
Hands-on120m
AgentCore Samples Repository - Deploy and Test
Install CLI, create Strands agent, deploy, test endpoint, check CloudWatch.
Study120m
Build Your Own Service Comparison Tables
From memory, build tables for: Bedrock vs SageMaker, Agents vs AgentCore, Vector stores, Caching, Guardrails filters.

Write your understanding, then reveal the reference answer.

0/4 reviewed

Build real muscle memory with these activities.

intermediate 60 min

Use the AgentCore CLI to create and deploy a Strands agent to production in minutes.

1 Install the AgentCore CLI: pip install agentcore-cli
2 Run agentcore create my-agent --template strands to scaffold a new agent project
3 Review the generated agent code: agent.py, tools.py, requirements.txt
4 Run agentcore dev to test locally with hot-reloading
5 Run agentcore deploy to deploy to AgentCore Runtime and note the endpoint URL

intermediate 45 min

Implement short-term session memory and long-term episodic memory in your deployed agent.

1 Open the AgentCore console and navigate to your deployed agent
2 Enable session memory and set the TTL to 1 hour
3 Send a series of messages: 'My name is Alice', then 'What is my name?' — verify session context
4 Enable long-term memory and configure it for episodic storage
5 In a new session, ask 'What do you remember about me?' to verify cross-session memory recall

20 questions across 3 difficulty levels.