Skip to content
VirtualOnlinePast event

LLM evaluation: A live demo

Event artwork for LLM evaluation: A live demo

This event has ended

This listing is kept as an archive. Explore upcoming events to find your next one.

See upcoming events

Supporting this event

Sponsors

Sauce LabsTUI
About this event

What to expect

Discover how to evaluate Large Language Models (LLMs) in test automation during this hands-on online demonstration exploring the challenges of testing non-deterministic AI systems.

What You'll Learn

As test automation engineers have traditionally relied on consistency and deterministic behaviour, testing LLM-powered applications presents a fundamental shift. This session tackles the unique challenge of evaluating software that is non-deterministic by design, featuring a live demonstration of real-world evaluation techniques.

Host Anupam Krishnamurthy will demonstrate:

Who Should Attend

This session is ideal for test automation engineers, QA professionals, and anyone working with AI-powered applications who needs practical strategies for evaluating LLM behaviour. Whether you're new to LLM testing or looking to strengthen your approach to AI quality assurance, you'll gain concrete techniques you can apply to your own projects.

Event Format

This interactive online session includes a 50-minute live demonstration followed by 15 minutes of Q&A and open discussion. You'll see the evaluation process in action, understand how to debug LLM behaviour, and explore security vulnerabilities specific to LLM applications.

The event is hosted by CTM: Continuous Testing Meetup London, bringing together 56 attendees from 4 testing and quality assurance communities. Sauce Labs and TUI sponsor this meetup series supporting the continuous testing community.

Register for this event

Running an AI event?

List it here, free.

Meetups, workshops and conferences go in front of thousands of builders after a quick review.

List your event