Generative AI for DevOps: Build Your Own LLM Server (English) (Virtueel)

Lesmethode :

Virtueel

Algemeen :

As a DevOps engineer, do you rely on ChatGPT but face restrictions at work due to company policy or sensitive data? Are you curious about hosting your own Large Language Model (LLM)?

In the one-day workshop Generative AI for DevOps Engineers, developed by the open-source experts at AT Computing and exclusively available via Vijfhart, you will learn how to set up and run your own local version of "ChatGPT". Everything is built on open-source technology without sending any data to the cloud.

By attending this course, you will directly meet the EU AI Act requirements on AI Literacy: risks, privacy, data sharing, model bias, and hallucinations are all thoroughly addressed. This ensures you not only gain hands-on skills, but also remain fully compliant with upcoming regulations.

The day begins with an introduction to Generative AI, exploring available LLMs and how to use them effectively. You will then dive into the hardware side, including GPU acceleration on virtual machines and in containers.

Through hands-on exercises, you will set up your own LLM server and even create a custom "model". You will also learn how to connect a web-based client to your LLM effectively building your own ChatGPT clone.

On top of that, you will explore how to integrate the LLM API with Python, apply Retrieval Augmented Generation (RAG) to work with your own documents, analyze images, and even perform log analysis using your LLM.

Doel :

After following this course, you will have knowledge of the internal workings of GenAI in general and LLM's in particular. You will also learn how to set up a local LLM server without sending your data to a Cloud environment. This course covers all topics for AI Literacy as required by the EU AI Act regulation.

Doelgroep :

The following prior knowledge is required:
- Linux Infrastructure
- Linux/UNIX Fundamentals
Knowledge of a programming language, preferably Python, is beneficial.

Voorkennis :

To follow the course, experience with the (Linux) command line interface is required, for example, Linux/UNIX Fundamentals or Linux Infrastructure. Knowledge of a programming language, preferably Python, is beneficial.

Onderwerpen :

- What is Generative AI?
- Which LLMs are available and how do you use them?
- Tokens, vectors, and parameters
- Hardware: CPU versus GPU: how to apply GPU acceleration on Virtual Machines and in containers (Docker)
- Setting up an LLM server
- Creating your own LLM model
- Applying a web-based client to your own LLM
- Connecting to the LLM API with Python
- Using your own documents with your own LLM
- Analyzing images with your own LLM
- Log analysis using your LLM