11. Hands-on LLM Ops: Setting Up Your Python Dev Environment and Project Structure

Key Takeaways

First let's set up the development environment and understand the structure of project. Most lag applications are implemented using Python. So the first step is creating an isolated virtual environment. This environment ensures that project dependencies remain separate from other Python projects on your machine. So let's start by creating an environment first. Okay. So we need to use this particular command to create a new environment. VNV is a is a command through that you can create a environment. So the environment name which I'm taking over here is LLM ops env. You can take any name any environment name. So if you can see a new environment has got created and and for that folder is also here. So any Python module we will install will get installed in this particular folder only. So let's first activate this particular environment. To do that the command is we need to go inside this particular folder. There is bin inside this. Then activate command using this we can activate this particular environment. Now we will see the folder structure. The if you can see over here this top folder lm ops code in this there are different folders the project is divided into the multiple modules okay if you can see this config file the configuration file which controls the behavior another file if you can see over here the core module it contains main orchestration logic one utility file file you will find over here yeah so in this this is basically handling document in JSON and pre-processing and the main file obviously it exposes the system as an API and the evaluator module evaluator module evaluator modules helps you to measure the response quality and the rag evaluator module which helps to measure response quality. So the key idea here is that the project is structured into the clear modules that separates configuration, system logic, injection and evaluation.