Machine learning models for churn and revenue prediction. Project Engagement: GitHub: https://gitlab.cee.redhat.com/pnt-ops/cloud-services/red-hat-cloud-services-machine-learning/rosa-machine-learning/rosa_growth
AI
Aspen social graph analysis
Understand changes in strategic open source ecosystems and identify new strategic ecosystems
ROSA – Foundation Models Generative AI for Doc Search
A useful application of foundation models, or large language models (LLMs), is documentation search. LLMs can expand the scope of product support, enabling users to access specific details and insights without the need to manually sift through voluminous documents....
Codeflare Stack Integration to ODH/RHODS
Enable machine learning for AI Developers with faster time to value. This project carries into the completed Ray on ODH work that was previously performed. There is additional alignment and integration that is on-going with the ODH and RHODS teams.
HCS – Red Hat Subscription Delivery
Increase productivity of Global Business Delivery (GBD) and increase revenue from Red Hat product subscriptions
ShadowBot.AI
Instant access to the company's internal data & documentation via a LLM-assisted chatbot that is integrated into tools Red Hatters already use. Project Engagement: GitHub:...
Open MLOps Stacks – Data Labeling
The team has started to work on the MLOps stack for Data Labeling. Both traditional machine learning training, and fine tuning of generative AI models, requires labeled data. The first goal for this KR is to conduct an ecosystem evaluation of open source labeling...
generative AI for CVE backports
Use generative AI code generation to automatically port CVE patches across multiple versions of Red Hat Products. Based on historical examples of applied patches, Red Hat ET will leverage or fine-tuning large code and language models developed at IBM Watson only...
Open ML-Ops Stacks – VectorDB
Study some open vector-db approaches - these will include things like postgressql+vector plugin but also purpose built tools like milvus. Evaluate for "ergonomics" and also performance. Performance may include throughput or retrieval metrics.
Confidential Compute for ML Workloads in OpenShift
The goal of this project is to understand the technical requirements and any gaps around being able to run ML Workloads in the data center while using Confidential Compute environments.