Instructlab

Red Hat’s InstructLab is a community-driven project that makes it easier for developers to experiment with IBM’s Granite models, for those with minimal machine learning experience.

​

It lets you experiment with the AI model of your choice or fine-tune foundation models on your local hardware. This removes the cost and resource barriers to experiment with AI models.

​

During my time as a Principal Data Scientist at Red Hat, I supported the development of synthetic data generation techniques for model tuning and built the CLI tooling supporting it. I was one of key members in founding team of InstructLab and served as a core maintainer. InstructLab’s model-agnostic approach simplifies the process of extending and refining LLMs, significantly lowering barriers for contributors and promoting collaborative AI advancement within the open-source community.