How do you test your application when the core functionality is powered by non-deterministic, generative AI? The rise of Large Language Models (LLMs) calls for a new unit testing paradigm. Can an AI assert its own behavior? Let's find out!
Development

Håkon Hapnes Strand is the CTO of Stavanger-based conversational AI startup Sklls. With a decade of experience in machine learning and data science, he has been doing AI since before it was cool. These days, he is busy building the next generation of voicebots at Sklls and embracing the generative AI revolution.
How do you test your application when the core functionality is powered by non-deterministic, generative AI? The rise of Large Language Models (LLMs) calls for a new unit testing paradigm. Can an AI assert its own behavior? Let's find out!