medium10 pts · first answer countsmodel-servingdata-validation
Silent wrong predictions from bad input
A client started sending a feature in cents instead of dollars and another as a string. The model kept returning predictions without errors. What should the serving layer do?
Discussion 0
Join free to post in the discussion.
Related questions
- When batch inference fitsMLOps Engineer
- Traffic before the model has loadedMLOps Engineer
- Why export a model to ONNXMLOps Engineer
- Bad data reached trainingMLOps Engineer
- Protocol for internal inference callsMLOps Engineer
- Low GPU use under loadMLOps Engineer
- p99 latency spikes during traffic burstsMLOps Engineer
- Hundreds of low-traffic modelsMLOps Engineer