Practically AI Leeds Launch Event

A celebration with pizza and presentations all about AI

This is the launch of something special; the popular Practically AI event is coming to Leeds from Manchester.

This event is the first event of many of this new group, which will continue to have regualar events for the forseeable future.

The evening will consist of a typical meetup format, with doors opening at 17:30 for arrival and a bit of socialising and food, and then at 18:30 an AI spotlight followed by 2 talks.

Talk One:

Prompt > Context > Harness > Loop: The Evolution of Agentic Software Engineering – Phil Whittaker

Loop engineering is the phrase of the moment, but getting started can feel like learning to run before you can walk.

In this talk, we’ll work through the stages most engineering teams experience as they move from simple prompting to managing context, building effective agent harnesses, and finally creating reliable autonomous development loops.

Along the way, we’ll also examine the traditional software engineering practices that make these systems work.

This talk will show that software engineering is not being replaced.

The tools are changing, the feedback loops are becoming faster, and automation levels are increasing – but designing, validating and operating reliable systems still requires engineering discipline.

This is not the end of software engineering.

This is software engineering.

Talk Two:

The Gap between AI Capability and Governance – Avu Harvey

AI can now do more than ever but our ability to control, trust and govern what they do hasn’t kept pace.

This talk focuses on the less glamorous, more critical layer underneath: Governance.

She’ll start the discussion with a recent example:

OpenAI disclosed that one of its test AI agents broke out of a controlled testing environment and autonomously hacked into the infrastructure of another AI company without any human instruction. This signals that AI capability is now outpacing our ability to govern it safely.

Then, she’ll dive into how governance (focused on access, oversight and accountability) is the real bottleneck for trustworthy AI. She will then unpack what “governance” needs to become in a world of autonomous AI agents beyond policies, the recommended practical controls, and oversight that keeps up with what AI systems are now capable of doing.

Finally, she will share resources for those interested in delving deeper into AI governance.