Software Engineering Radio - The Podcast for Professional Software Developers · SE-Radio Team
Jacob Visovatti and Conner Goodrum on Testing ML Models for Enterprise Products
·1 hr 1 min·5 clips
Jacob reveals how AI models now handle unstructured data in real-time, exposing critical quality concerns for enterprises.
As heard by us
A practical look at why ML evaluation gets harder once voice models have to serve enterprise products.
Testing ML models for enterprise use cases is framed as a measurement problem rather than a slogan. Conor Goodrum and Jacob Biswati focus on the gap between standard metrics and real product behavior, especially in voice technology, where a model can look strong on paper and…
Why you'd press play
If you ship ML into an enterprise product, this episode shows where headline metrics stop telling the whole story.
Listen to the show on