Test Workflows and Generative AI
summary
Claims that generative AI draws over $100 billion in investment and that models now have billions to trillions of parameters. LabVIEW gets a built assistant called Nigel that uses Azure fine‑tuned LLMs, a vector DB and RAG retrieval to generate and document VIs, suggest fixes, perform semantic searches and reference hardware interfaces; it runs in a tech‑preview with NDA, requires internet, tracks token costs, uses OpenAI human eval and pass@k to measure code quality, and still hallucinates and needs human oversight.
AI-written summary · about
chapters (9)
- 0:00Introduction
- 4:11Generative AI history
- 11:30Generative AI adoption
- 18:41Generative AI in test workflows
- 24:40NI’s responsible AI policies
- 27:28Demo: NI’s AI assistant in LabVIEW
- 34:39How AI models understand responses
- 42:04AI challenges and limitations
- 45:55Future explorations
related talks
49mGenerative AI: Concepts, Tools, and Applications for Data Professionals
20mGenAI Application Development
41mThe future of AI and generative code Q&A
29mImpact of #GenerativeAI in Development and #SoftwareTesting
40mDavid Fogl: Leveraging Generative AI for Enhanced Software Testing
41mGenerative AI: bliss or miss?
40mCode is Life - Building a Responsible Generative AI Solution
41mAI-Augmented Testing: How Generative AI, Prompt Eng Turn Testers Into Superheroes