Course Objective & Scope
ReAct is the pattern in which an LLM interleaves reasoning steps with tool calls, and it sits underneath most agent frameworks. This course shows how the loop works by coding it directly against the OpenAI, Anthropic Claude or Gemini APIs, without a framework.
Participants study observation formatting, iteration caps, infinite loops and malformed tool calls through live traces. The concrete output is a working, tool-using ReAct agent on a use case chosen by the organisation, plus a catalogue of its failure scenarios.
Who Should Attend?
- Software developers building LLM-based applications
- AI engineers prototyping agents
- Data scientists who want to understand what frameworks do under the hood
- Prerequisite: Python and working with REST APIs
Key Takeaways
- Implement the ReAct loop independently of any framework
- Write tool definitions in JSON Schema and wire them to native tool calling
- Prevent runaway loops with stop conditions and iteration limits
- Analyse agent decisions through traces and classify failures
- Judge which tasks ReAct suits and where it falls short
Curriculum
The Logic and Evolution of ReAct
- From chain-of-thought to ReAct: combining reasoning and acting
- Text-based ReAct versus native tool calling
- The changing role of ReAct with reasoning models
Coding the Loop from Scratch
- Message history, tool calls and appending observations
- Formatting tool results for the LLM
- A first working agent with a simple tool set
Stop Conditions and Control
- Maximum iterations, token budgets and timeouts
- Detecting the final answer and structured output
- Catching repeated actions and loops
Failure Modes and Debugging
- Hallucinated tools and malformed parameters
- Returning tool errors to the agent as meaningful messages
- Reading traces and classifying failures
Applying It to an Organisational Use Case
- Designing a tool set for a real business task
- Measuring success rate with a small eval set
- The limits of ReAct and criteria for moving to other patterns
Other Courses in This Category
-
Technical course on agents that split multi-step tasks into planner, executor and replanner roles, covering ReWOO and LLMCompiler-style parallel task graphs, plan validation and cost optimisation.
- Live Course
- 5 modules
-
Course for teams building long-running agents: state schemas, reducers, checkpoints, thread management and state-machine workflows, illustrated with LangGraph and workflow engine examples.
- Live Course
- 5 modules
-
Hands-on course on designing tools agents can use reliably: naming, descriptions, JSON Schema parameters, error messages, tool contracts, versioning and permission scopes.
- Live Course
- 5 modules
-
Hands-on technical course on building production-ready agent systems with LangGraph: StateGraph, conditional edges, checkpointers, interrupt-based human-in-the-loop, subgraphs and the Send API.
- Live Course
- 6 modules
-
Course on building tool-based agents with the OpenAI Agents SDK and Responses API: function tools, hosted tools such as web search and file search, handoffs, guardrails, sessions and built-in tracing.
- Live Course
- 5 modules
-
Hands-on course on building Model Context Protocol servers that expose tools, resources and prompts, and clients that consume them: stdio and Streamable HTTP, OAuth-based authorisation, security and testing.
- Live Course
- 5 modules
Get in touch
