We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
David: Making MLOps uncool again
Expert MLOps speaker David explores simplifying the process for researchers by showing how to automate workflows with GitHub Actions and DVC, highlighting limitations and importance of seamless experiences in streamlining research workflows.
- David’s workshop aims to show that MLOps is still in its early stages and needs improvements to make it more user-friendly for researchers.
- MLOps solutions like Kubeflow are good but have limitations and may not provide a seamless experience for researchers who want to focus on their work, rather than dealing with complex DevOps and ML tooling.
- The goal of making MLOps uncool again is to focus on simplicity and ease of use, making it accessible to researchers without requiring extensive DevOps knowledge.
- Researchers should be able to work directly with their data, models, and results without having to worry about setting up complex infrastructure or maintaining their own environments.
- There are many MLOps solutions available, but they often require a good understanding of DevOps and may not provide the necessary level of customization or flexibility.
- David’s workshop aims to showcase how to set up an automated issue labeler using GitHub Actions and DVC, making it easier for researchers to automate their workflows and focus on their research.
- The workshop also highlights the limitations of existing MLOps solutions and the importance of providing a seamless experience for researchers to simplify their workflows.
- Researchers should not have to worry about setting up complex infrastructure or maintaining their own environments, but rather should be able to focus on their research and experimentation.
- David’s workshop aims to showcase a simpler approach to MLOps, one that is more accessible to researchers and less dependent on complex DevOps tooling.
- The workshop uses a real-life example of an automatic issue labeler and showcases how to set up the necessary workflows and infrastructure using GitHub Actions and DVC.
- The goal is to demonstrate that MLOps can be more accessible and user-friendly, making it easier for researchers to automate their workflows and focus on their research.
- The workshop also highlights the importance of providing a seamless experience for researchers, one that is free from unnecessary complexity and allows them to focus on their research and experimentation.
- Researchers should not have to worry about setting up complex infrastructure or maintaining their own environments, but rather should be able to focus on their research and experimentation.