Agentic AI: Conversational Agents

In this course, you'll learn:

Agentic AI Primer (Core Concepts & Terminology)

• Defining agentic systems and how they differ from standard automation
• Agentic components: memory, tools, and reasoning loops
• Comparing agentic workflows with RAG and fine-tuning
• Shared terms: planner, executor, environment, guardrails 

Anatomy of a Conversational Agent

• Conversational agents vs. chatbots
• Key LangChain components and agent workflows
• Creating a minimal  LLMChain

Prompt Engineering for Dialogue

• System and user prompt design
• Tone matching and safety phrasing
• Few-shot prompting and verbosity control

Conversational Memory

• Short-term vs. summary memory 
• Privacy tradeoffs and memory configs
• How memory impacts response quality

Vector Search & Document Embedding

• Embeddings and vector search basics
• Using FAISS and chunking strategies
• Document retrieval with OpenAI models

Retrieval-Augmented Generation (RAG)

• How RAG improves accuracy and grounding
• Using LLamaIndex with LangChain
• When to prefer RAG over direct embedding

External Tool Integration & Function Calling

• Tools vs. APIs in LangChain
• Function calling and controlled execution
• Routing inputs and chaining results

Evaluations & Guardrails

• Latency targets for user experience
• Token usage monitoring and cost control
• Content filtering and fallback design

Deploying via Hugging Face Spaces

• Deployment steps and required files
• Streamlit vs. Gradio UI basics
• Secret and token management