How to build autonomous AI agents

A practical session on building autonomous agents that can react to real-world triggers, operate in the cloud, and verify their own work before handing control back to a human.

Mar 26, 2026 · UCD O'Connor Centre for Learning, Belfield, Dublin

The Augmented Software Engineer — Agentic-Coding - Practitioners Day (II)

Presentation

Talk overview

  • Using operational triggers like new Sentry issues, incoming user emails, and pull requests to kick off autonomous workflows.
  • Designing agent loops that can plan, execute, and check their own work instead of stopping at a single prompt-response cycle.
  • Hardening long-running agents with validation, monitoring, and clear human checkpoints before production actions land.