How to effectively test your AI applications?
summary
How to test AI-driven applications effectively by addressing challenges like probabilistic outputs, exploratory workflows, and changing contexts in ML development. Covers software development processes and practices that Data Scientists and Machine Learning Engineers can use to make ML systems reliable and maintainable while retaining the ability to quickly experiment and adapt models for data drift or shifting functional requirements.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 4:36Equinoxe and Mouvance
- 11:13Data exploration and preprocessing
- 22:35ML model training
- 30:58Pipeline integration and testing
- 38:38Deployment and A/B testing
- 44:31Model maintenance and monitoring
- 46:18Q&A
related talks
45mHow to Test Artificial Intelligence and Machine Learning Solutions?
32mQuality engineering in the AI age: From testing AI to leveraging GenAI
45mTesting Applications That Use AI
01h 04mWhat Testers Need from Machine Learning and Artificial Intelligence
44mApplying AI Capabilities in Scenario Testing
07m“AI ML in Test Automation, Are you ready?”
44mHow AI is Exposing the Existential Crisis of Software Testing
44mAI workflows and teaming - how do teams deliver AI products?