WEEK 6
AI Agents
An agent that completes a multi-step task alone.
Course Details
Students transition from building AI assistants to creating autonomous AI agents capable of planning, reasoning, using tools, and managing their own execution state. They learn how agent loops work, when to use workflows versus agents, implement ReAct reasoning patterns, and build production-style agents that complete complex multi-step tasks with minimal human intervention.
What you will learn :
Workflows vs agents, and when to use each
The agent loop, built from scratch
ReAct, reflection & planning patterns
Agent state management
Stop conditions & when NOT to use an agent
Topics Covered
Python (agent loop)
OpenAI / Anthropic Tool Use
LangGraph (intro)
State management
Project 6
OpsAssist · ResearchAgent
Project : OpsAssist
Project Description
Build an autonomous operations agent that accepts high-level requests, plans the required steps, selects appropriate tools, executes tasks sequentially, tracks its internal state, and determines when the objective has been completed. Students implement a complete agent loop with planning, reasoning, tool use, and stop conditions.
Project Result
Developed a production-style AI operations agent capable of autonomously planning and executing multi-step workflows. Implemented agent state management, ReAct reasoning, dynamic tool selection, and intelligent task completion using Python and modern LLM tool-calling capabilities.
Project 2: ResearchAgen
Project Description
Build an autonomous research agent that investigates a topic, gathers information from multiple sources, iteratively refines its search strategy, and compiles a structured research brief. Students implement planning, reflection, tool use, and state tracking to enable end-to-end autonomous research.
Project Result
Created an intelligent research agent that independently plans research tasks, retrieves and analyses information, synthesises findings, and generates structured research reports. Demonstrated autonomous reasoning, iterative planning, tool orchestration, and agent state management in a production-style AI workflow.
Back to Home