Forward Deployed Architect Program in Cambodia

Module 1: Overview of Generative AI and LLMs (4 Hrs)

• Artificial Intelligence and Machine Learning
• Key types of Machine Learning- Supervised & Unsupervised Learning
• Basics of Deep Neural Network
• Natural Language Processing & Generative AI
• Overview of LLMs & Multimodal LLMs
• Demo - Solving ML, NLP Problems using LLMs

Hands-On Activities / Labs

• Multimodal Embedding
• Subword tokenization

Module 2: Working with Retrieval Augmented Generation (6 hours)

• Concept of Retrieval Augmented Generation (RAG) with 2 real-life examples
• Data pipeline for RAG
• Demo/ Code walkthrough- Naïve RAG
• Breakout Room Activity- Understanding code/ flow of an E2E RAG pipeline
• Business use cases of RAG Apps across Industries and Horizontals/ Functions, including Vertical AI Agents
• Chunking Strategies
• Creating Vector embeddings and storing them in a Vector database, with examples
• Retrieval Mechanics & Similarity/ Vector search techniques
• Langchain for RAG Development
• Techniques for improving RAG accuracy, Reranking, chunk optimization, and others
• Langflow for low-code, no-code visual modeling-based RAG development
• Lab: Building RAG Data ingestion, indexing, retrieval, and generation pipeline
• Advanced RAG techniques- Agentic RAG, GraphRAG, Multimodal RAG

Hands-On Activities / Labs

• Agentic RAG
• Multimodal RAG
• Graph RAG
• RAG Evaluation

Module 3: AI Agents & MCP Servers- Deepdive (20 hours)

• Overview of AI Agents/ Agentic AI & role of LLMs/ SLMs
• Tool use/ function calling
• MCP Servers - connecting AI Agents to 3rd party apps
• MCP Client & MCP Host
• Business use cases of AI Agents across Industries and Horizontals/ Functions, including Vertical AI Agents
• Technical use cases of AI Agents (in SDLC automation)
• MCP Server Security
• Step-by-step - Designing an AI Agent
• Building Agent with tools & MCP Server - Google ADK, AWS Strands SDK, Langgraph
• Advanced Context Engineering for Agents
• Agentic Memory - Long-term, short-term, with demos
• Agents controlling browser, keystrokes, and mouse clicks/ computer control with demos (short)
• Deep Agents- working with complex, long-running tasks
• AI Agent Building Tools- Code-based and low-code, no-code
• AI Agent/ Agentic AI Pricing models & how they are disruptive to existing SAAS models
• Breakout Room Activity: Building use case and specs for an AI Agent
• Multi-agent systems
• Quiz and Q&A

Module 4: Building & integrating AI Agents- Technical Deepdive (38 hours)

• Demo- Building a Multi-agent system with Langgraph
• Demo- Building a custom MCP Server with FastMCP
• Demo- Building an AI Agent with AWS Strands SDK
• Demo- Building an AI Agent with Langflow
• LLM/SLM selection for AI Agents/ Agentic AI
• Demo: Reflection in AI Agents
• Building multi-agent systems
• Google Agent2Agent (A2A) Protocol for multi-agent collaboration with demo
• Demo: Multi-agent collaboration 
• Agentic RAG
• Agent Skills and Agents.md
• Designing and building Agent UI-UX / Conversational experiences - Good practices, frameworks & tooling
• Agent Harness
• Different ways of training Agents on custom enterprise data
• Breakout Room Activities - Technical Design of an AI Agent 
• Quiz and Q&A

Hands-On Activities / Labs

• Build a multi-agent system with Langgraph (Select domain use cases)
• Build a multi-agent system with AWS Strands SDK
• Build a multi-agent system with Langflow (Select domain use cases)
• Context Engineering- 3 labs
• Agentic Memory- 2 labs
• Reflection in AI Agents
• Multi-agent collaboration with A2A protocol
• Agent Skills (Select domain use cases)
• Agents.md (Select domain use cases)
• Use GitHub Copilot Agent mode & connect with GitHub MCP Server 
• Build an AI Agent with MCP Server Integration
• Agent Harness

Module 5: Production deployment of AI Agents

• Responsible AI and AI Governance for Agents (Databricks AI Governance Fwk, ISO42001, NIST RMF)
• Agent Evaluation
• Select customer case studies on Agent and RAG Evaluation
• Agent Security (Agent identity, Threat modeling, OWASP Top 10, Red teaming)
• Agent deployment & scaling options on AWS Bedrock, Azure AI Foundry, Google Enterprise Agent platform
• Post-production considerations for AI Agents
• Demo - Guardailing LLM/ Agent
• Building well-architected agents/ Architectural perspectives
• How to optimize Agent cost/how to build solutions with token constraints
• Quiz and Q&A

Hands-On Activities / Labs

• LLM-as-a-judge for Agent Evaluation
• Agent Evaluation with Galileo AI