Build and deploy LLM-based applications
LLMs have gained immense popularity in recent months. An entirely new ecosystem of pre-trained models and tools has emerged that streamline the process of building LLM-based applications.
Join us for this in-person workshop in San Francisco, CA to experience building scalable LLM-based applications hands-on! Coffee and snacks included. This is a paid workshop to help cover cloud resources and operational costs.
Key Takeaways
- Use libraries like Ray, HuggingFace, and LangChain to build LLM-based applications based on open-source code, models, and data.
- Learn about scaling LLM fine-tuning and inference, along with trade-offs.
- Use embedding models and vector stores.
- Learn about using modern deployment tools to run your application online and continually improve it.
Audience
- This workshop is for LLM-based application builders as well as ML practitioners interested in how to build LLM-based applications using open source tools.
- Basic understanding of emerging LLM trends is not necessary, but will help get more out of the workshop. This and this blog posts are helpful background material.
Resources
- GitHub repository with relevant resources including notebooks, setup instructions, bonus resources, and a README for an overview.
- Access to the GPU-based compute cluster for the duration of the training to leverage the full potential of LLMs.
- Exclusive only to the workshop participants: office hours the week after the workshop for deeper discussion of the individual application ideas.
![[San Francisco Workshop] Deploy and Scale LLM-based applications](https://cdn.gradual.com/images/https://d2xo500swnpgl1.cloudfront.net/uploads/mlops/Copy-of-LLMs-Prod-Part-II-slides-7--9fdb9642-3a31-4c49-96a0-a65ef91eb28b-1686330034584.png?fit=scale-down&width=1200)
