← Financial Cloud Cloud Cloud Club · Amarathon 2025 Recap

Vertex Macro|Financial Cloud Cloud · AWS Amarathon 2025

Vertex Macro|Amarathon 2025 Recap 06:From Matching to Understanding: Personalized AI Search Practice Driven by AgentCore Memory

Speaker: Liu Cao

Session: 06

Session
Summit Dev Lounge2026 Re:cap
01 Encode Architecture as Steering for AI Agents
02 Agent Harness Is the Real Engineering Moat
03 Ask Observability Data in Plain Language
04 Serverless AR Game with Bedrock AgentCore
05 Multi-Agent Quant Backtesting on AgentCore
06 Blog to Slides in Three Minutes with Kiro
AWS Community Day Hong Kong 2025 Re:cap
02 AWS Compliance with Terraform
03 Beginner to Builder An AWeSome Cloud Journey
04 Team-First Serverless Engineering with Laravel & Bref
05 Event Opening - AWS Community Day Hong Kong 2025
06 Agent-to-Agent: Building Interoperable AI on AWS
07 Utilize another telemetry data for faster improvement with AI agent
08 Graduating from Vibe Coding: Spec-Driven Development with Kiro
09 Automated Testing using MCP & AI Agents
10 Modernizing Telecom Security ML Powered Approach
11 Rethinking GenAI Agent: RAG & MCP
12 Disaster and Emergency Response with TAK and AWS
13 Rethinking Serverless Application Workflows from a Testing Perspective
14 Practical AWS FinOps for Cloud Success
15 AI-Powered Global Pure-Alpha Macro Trades on AWS: Revolutionizing Risk-Adjusted Asset Returns
FSI Recap
01 Modern Trade Lifecycle: Trading to Settlement
02 Goldman Sachs: Fast Track your applications onto Cloud - AWS Re:cap Q1/2023
03 Zurich Insurance Group: Building an Effective Log Management Solution on AWS
04 FSI Meetup 2025 Q4 - Brex Database Disaster Recovery
05 FSI Meetup 2025 Q4 - A Graviton Migration Success Story
06 FSI Meetup 2025 Q4 - Stifel Modern Data Platform
07 FSI Meetup 2025 Q4 - Financial Transaction Data Reconciler PayPal
08 FSI Meetup 2025 Q4 - Scaling Resilience
09 Maximizing AI Inference Cost Efficiency: Strategic Adoption of AWS GPU Instances
10 Advanced Agentic AI Design Patterns
11 Build New Modern Apps on AWS
AWS re:Invent 2025
01 Coinbase re:Invent Recap (IND3312)
02 Building the Future Trading Platform Leveraging AI and AWS
03 Trading Innovation: Jefferies' AI Assistant on Amazon Bedrock (IND3315)
04 How FSI Revolutionized HFT Analytics with Agentic AI (GBL302)
05 Improving Distributed Systems with Amazon Time Sync Featuring Nasdaq
06 Amazon Aurora HA and DR Design Patterns for Global Resilience (DAT442)
07 Building Agentic AI: Amazon Nova Act and Strands Agents in Practice (DEV327)
08 Deep Dive into Amazon Aurora and Its Innovations (DAT441)
09 Deep dive on Amazon S3 (STG407)
10 Nasdaq: Build Resilient Infrastructure for Global Financial Services (HMC327)
11 What's New with AWS Lambda (CNS376)
12 Spec-Driven Development with Kiro (DEV314)
13 Amazon's finops: Cloud cost lessons from a global e-commerce giant (AMZ308)
14 Tick to trade latency trading platforms on aws
Government data
01 The AI Era: The Boundary Between Development and Design Is Disappearing
02 On-Device Multimodal AI and Smart-City Practice
03 Large-Model Capability Evaluation and a Method for Landing AI Projects
04 Controlled End-to-End Automation of Government Development with Cloud Agents
05 A New Software Ecosystem for the Agent Era, Seen Through Multi-Agent Systems
06 AI-Driven Macro Quantitative Research and Smart Governance
07 Authorized Operation of Public Data and Smart-Government Practice
08 Putting Data Assetization into Practice: Rights, Compliance, Engineering Governance, and Digital-Government Cases
09 AI for Mental-Health Public Welfare: Governance, Architecture, and Practice of a Trustworthy Platform
Amarathon 2025 Recap
01 A Developer’s Roadmap to Architecting for Agents
02 Amazon Bedrock Data Automation
03 Multi-Agent on AgentCore
04 Building Agentic AI Nova Act and Strands Agents in Practice
04 Accelerating Migration Projects with Kiro using Spec-Driven Development
06 From Matching to Understanding: Personalized AI Search Practice Driven by AgentCore Memory
07 Observe to Optimize – LLM Observability to AIOps Turning real-time insights into intelligent automation
08 Deploying TEAM and Building the Best Engineering Team
09 Five Hard Lessons from Five Years of So-Called Serverless Databases
14 What if AI does my job How Q Developer CLI and Kiro have changed my daily routine
16 Velocity with Vigilance: Security Essentials for Amazon Bedrock Agent Development
26 Run OSS LLMs on a Single H100 Smarter, Cheaper, Faster
28 A Modern Unified Metadata Architecture: New Approaches to Breaking Down Data Silos
29 Serverless MediaOps: Automating Video Workflows with AI on Amazon Web Services
30 Architecting for Efficiency and Reliability with Performance Testing at Scale
31 Connecting the World Through Open Source: Practical Journey of Technology, Community and Global Developer Relations
33 Building Streaming Iceberg Tables for Real-Time Logistics Analytics
34 Accelerating Large-Scale Robot Strategy Training: An Automated Closed-Loop Architecture Based on Kiro, Trainium, and EKS
35 From Vibe to Viable with spec driven development
36 Making Cloud Cost Analysis Smarter: Building FinOps Intelligent Agents with Strands and AgentCore
37 Transform Conversational Agentic AIOps for K8s Using CNCF Kagent, K8sGPT, and Nova Sonic

Search

● Search Engine Architecture Flow

Input

● Query

Retrieval Methods (from Query)

● Inverted index based lexical matching

● Item-based collaborative filtering

● Embedding-based retrieval

Merge Stage: Combines outputs from the three retrieval methods into

● Unordered product set without duplication

Ranking Stages

● Pre-ranking

● Relevance ranking

● Ranking

Additional Inputs to Mix-Ranking

● Advertising

● Content media

Final Ranking Stage

● Mix-ranking


System architecture for augmenting product matching using semantic matching

Input

● Query

Matching Components

● Behavioral Data

● Keywords Matching

● Semantic Matching

Data Flow

● Query feeds into Keywords Matching and Semantic Matching

● Behavioral Data provides a behavioral signal to Ranking

● Keywords Matching and Behavioral Data form an unordered match set

● Semantic Matching provides a semantic similarity score to Ranking

Processing Stage

● Ranking

Output

● Ordered product list


LLM Agent Workflow

Process Flow

● Query -->LLM -->Plan

● Plan -->LLM -->Actions

● Actions -->"Is the plan complete?" (Decision Point)

● If "yes" -->LLM -->Finalize answer

● If "no" -->Loop back to "Actions" (via an LLM step)

Decision Logic

● Uses an LLM to determine if the plan is complete.

● Differences

01 Traditional Search

● Relies on keyword matching

● Webpage link weight ranking

● Goal is to quickly retrieve massive information

02 AI Search

● Semantic understanding

● Related information injected into vectors + keywords

● Large models assemble information

● Goal is to accurately answer user questions


Memory

● LLM with Context Storage Architecture

Main Components

● User

● Chat APP

● LLM (Stateless)

● Context Storage

Workflow

● Send message (from User to Chat APP)

● Carry complete context (from Chat APP to LLM)

● Return response (from LLM to Chat APP)

● Store history (from Chat APP to Context Storage)

● Return result (from Chat APP to User)

Context Storage Content

● System Prompt

● Round 1: User Question

● Round 1: AI Answer

● Round 2: User Question

● Round 2: AI Answer


Architecture for an agentic system

● uses LLMs to manage memories and knowledge graphs from conversations:

Data Flow & Extraction

● Input: Conversations (messages) are fed into the system.

● Extraction LLM: A Large Language Model processes the messages to generate structured data.

● Outputs: "New memories" and "new entities & relations".

Storage Components

● Vector Database: Stores "existing + new memories" for retrieval and updates.

● Graph Database: Stores "existing + new entities & relations" for updates.

Update & Management

● Update LLM: A second LLM manages the data updates back into the databases.

● Functions: Manages "store updates" for both databases.

● Operations: Includes explicit "Add", "Delete", and "Update" actions for the stored data.

Overall Goal

● Memories ADD: The entire system facilitates the persistent storage and management of conversation context and extracted knowledge.


Enhanced Context Storage

System Components & Definitions

● System Prompt: System instructions or configuration.

● Dialog History: Record of the conversation.

● Tool Definitions: Available functions for the AI agent.

● weather_api(): requires location, date

● calculator(): requires expression

● Thinking: The agent's reasoning process.

Memory Interaction

● Extraction (from Thinking to Memory)

● Backfill (from Memory to Thinking)

● Memory: Long-term or short-term knowledge base.

Example Workflow: "Query weather (Fahrenheit)"

● User: Query weather (Fahrenheit)

● AI: Need to call weather API

● Tool Usage History:

● Call: weather_api(location='Beijing', date='today')

● Return: {'temperature': 25, 'condition': 'sunny'}

● Call: calculator(expression='25*1.8+32')

● Return: 77

Benefit

● Long-term retention and efficient management

● Continuous knowledge update

● Personalized service

● Complex task support

● Improve interaction quality

● From stateless to stateful


AgentCore Memory

Amazon Bedrock AgentCore

● Utilize any framework and model without managing infrastructure, safely build, deploy, and operate high-performance Agents at scale.

Components

● Runtime

● Memory

● Identity

● Gateway

● Code Interpreter

● Browser Tool

● Observability

Simplified Memory System Management

● Abstract memory infrastructure

● Based on serverless architecture for automatic scaling

● Automatically store and manage the context information of Agents across multiple sessions

Enterprise-Level Services

● Provide dedicated storage for each customer to fully guarantee data privacy

● Provide encryption protection and regionalized data storage to meet enterprise-level security needs

Deep Customization

● Customize memory modes according to specific application scenarios

● Set long-term memory extraction rules

● Select the appropriate model and customize the prompt words to optimize the long-term memory extraction effect


Agent Memory Components

Agent Core

● Agent Reasoning

● Agent State

● Knowledge Components

● Tool Calling

● Policy Definition

Short-Term Memory

● Context Window

● Conversation History

● Tool Calling History

Long-Term Memory

● Events Summary

● User Profile Information

● Document Information

Connections

● Storage - between Agent Core and Memory components

● Retrieve - between Memory components and Automatic Memory Retrieval Module

● Automatic Memory Retrieval Module


Agent Architecture & Benefits with Amazon Bedrock AgentCore Memory

Agent Functionality

● The intelligent agent automatically breaks down complex user needs (e.g., "Recommend a laptop under 5000 yuan for a university student that runs design software") into multiple parallel sub-tasks.

● Examples of sub-tasks: "University student usage scenario analysis", "Budget filtering", "Software running requirement matching".

Benefits of AgentCore Memory

● Relies on the technical advantages of AgentCore Memory's long-term memory.

● Significantly speeds up the Agent development process.

● Achieves multiple real-world application values:

● Token usage for smart searches based on user profiles is greatly reduced.

● Search result accuracy rate is effectively improved.

● Provides a massive breakthrough in technical competitiveness and cost-effectiveness.

System Diagram Flow

● User Query leads to Agent Runtime Environment (Strands Agents).

Strands Agents interacts with

● Tools (User Preference Retrieval)

● Amazon Bedrock LLM (Processes Output)

● AgentCore Memory

AgentCore Memory manages

● Short Term Memory (Interaction Events)

● Automatic Memory Extraction

● Long Term Memory (User Preferences)

● The process results in an Agent Response to the user.