FIT3224 · Engineering agentic systems

Overview

Engineering Agentic Systems introduces students to the principles, architectures, and engineering practices used to design and develop modern agentic software systems. The unit explores how intelligent agents interact with environments, tools, data sources, and other software components to perform goal-directed tasks in dynamic and distributed contexts. Students will investigate the structure and behaviour of agentic systems, including orchestration mechanisms, memory models, decision-making processes, and integration with external services such as APIs and databases. Through practical and project-based activities, students will analyse problem domains, design agent-based solutions, and develop software components that support autonomous or semi-autonomous behaviour. The unit also examines methods for testing, monitoring, and evaluating the correctness, robustness, and efficiency of agentic systems, while considering reliability, scalability, ethical implications, and potential system failures. Emphasis is placed on applying contemporary software engineering practices to the development and deployment of secure, maintainable, and effective agentic applications.

Offerings

The Handbook publishes no offerings for this unit.

Assessment

The Handbook publishes no assessment items for this unit yet.

Assessment details may change. Please refer to the assessment information in Moodle closer to the start of the teaching period.

Requisites

Learning outcomes

1. Explain the structure, components, and operation of agentic systems, including agents, environments, tools, memory, and orchestration mechanisms;
2. Analyse problem domains and interpret requirements to design appropriate agent-based software solutions and interaction patterns;
3. Develop software components that demonstrate goal-directed behaviour, decision-making, and interaction with external services and data sources;
4. Apply software engineering techniques to integrate agents with APIs, databases, and distributed software components to support reliable system operation;
5. Use testing, monitoring, and evaluation methods to assess the correctness, robustness, and efficiency of agentic systems in dynamic environments;
6. Evaluate trade-offs, constraints, risks, and potential failure modes associated with the design and deployment of agentic systems.

Workload

No workload detail published.

Student experience, not official rules. Nothing here changes what the Handbook says.