Presented by
lythelabs
simulated users at scale
2 Going
a tester's guide to llms, tools and agents
About Event
AI tools don't give the same answer twice, so normal tests can't say if they're ready to ship. Evals can. This talk shows how to find what goes wrong, turn it into test cases, and check the results in simple ways. It covers single answers, tools and agents. You'll see that the skills you already use as a tester carry over to AI.
After this talk, you will be able to:
Find what goes wrong in AI answers and turn it into test cases.
Choose simple checks and run them enough times to trust the results.
Test tools and agents step by step, using the testing skills you already have.
Hosted by LytheLabs & Taqelah
Sponsored by Superteams SG
Presented by
lythelabs
simulated users at scale
2 Going