Generative & Knowledge AI
Lifecycle stages:
MLOps journey description: This journey involves AI models capable of generating text, images, code, or recommendations, enhancing automation and personalization. It includes large language models (LLMs), chatbots, and deep learning-based recommendation engines.
Use case examples: Language, code, and content generation, as well as knowledge management using LLMs and generative models.
- Generating first-draft technical reports or maintenance logs
- Chatbots for internal technical support or field ops Q&A
- Summarizing long drilling or production logs
- Generative design of new components (e.g., drill bit designs)
- Knowledge retrieval assistants trained on internal documents, safety manuals, and regulatory texts
Related success stories:
- CSSU-OPT-KAI: Enables operating assets to improve risk management and deliver more reliable plans to the offshore organization by utilizing Natural Language Processing (NLP) models.
- KAI Enablers: An event-driven framework for serving Knowledge AI (KAI) models, helping users extract insights from unstructured data to enhance decision-making and solve business challenges.
- MATE Notifications: The Maintenance Analysis Tool for Equinor (MATE) processes machine failure logs written by maintenance engineers to predict failure modes and classes.
- Multilingual Models KAI: The multilingual models are trained and fine-tuned Brazilian-Portuguese ML models for equipment entity extraction. These models are used by Operational Planning Tools (OPT) at Equinor.
- Stable Diffusion Gen AI: A web-based Stable Diffusion experiment enabling users to create images from text prompts.