Course Objective & Scope
An agent can only do real work if it calls the right tool with the right arguments. The programme covers tool calling end to end: tool definitions with JSON Schema, how tool descriptions shape model behaviour, parallel calls, structured output and feeding results back into the context.
Using Python with the OpenAI, Anthropic Claude and Google Gemini APIs, participants build an agent with tools for CRM lookup, document search and record creation, test failure and timeout scenarios, and produce a reusable tool design checklist.
Who Should Attend?
- Software developers building LLM applications
- Integration and backend teams opening internal systems to AI
- AI engineers preparing agent prototypes
- Prerequisite: basic Python and REST API knowledge
Key Takeaways
- Write tool schemas and descriptions that models interpret correctly
- Implement the tool calling flow across different LLM providers
- Validate tool arguments and handle faulty calls safely
- Feed tool results back into the context efficiently
- Verify a multi-tool agent with test scenarios
Curriculum
How Tool Calling Works
- How the model selects and invokes tools
- Differences across the OpenAI, Claude and Gemini APIs
Tool Schema and Description Design
- JSON Schema, required fields and enums
- How tool naming and descriptions affect accuracy
Structured Output and Parallel Calls
- Producing schema-conformant responses
- Managing parallel and sequential tool calls
Connecting to Internal Systems
- REST API, SQL and document search tools
- Credentials and the principle of least privilege
Error Handling and Resilience
- Invalid arguments, timeouts and retries
- Reporting tool errors meaningfully back to the model
Lab: A Multi-Tool Agent
- End-to-end agent with CRM, search and record tools
- Test scenarios and tool design checklist
Other Courses in This Category
-
Method for turning business processes into agentic workflows: separating deterministic steps from LLM decisions, prompt chaining, routing, parallelisation and evaluator–optimizer patterns, implemented in n8n or Power Automate.
- Live Course
- 6 modules
-
Developer programme on building autonomous agents with LangGraph, the OpenAI Agents SDK and the Claude Agent SDK: state management, tool integration, loop control, tracing and a deployment-ready agent for an organisational case.
- Live Course
- 6 modules
-
Technical programme on how agents plan and execute multi-step tasks: ReAct, plan-and-execute, task decomposition, reflection and replanning, using reasoning models, and loop termination strategies.
- Live Course
- 6 modules
-
Agent memory and context management for long-horizon tasks and multi-session interactions: working memory, episodic and semantic long-term memory, context compaction, and persistent memory architecture with Mem0 and LangGraph Store.
- Live Course
- 5 modules
-
Multi-agent architectures that split complex work across specialised agents: supervisor, hierarchical and swarm patterns, comparing LangGraph, CrewAI and AutoGen, the A2A protocol and shared state management.
- Live Course
- 5 modules
-
Human-in-the-loop agent architecture that preserves human oversight on critical actions: interrupt/resume flows, risk-based approval thresholds, escalation queues and agents that improve from human feedback.
- Live Course
- 5 modules
Get in touch
