Check your environment

Outcome: choose an installation route without guessing which terminal or machine a command targets.

Supported scope

The main development environment is Windows with WSL2 Ubuntu. The manual lab walkthrough starts the backend in a WSL terminal. Existing systemd deployments commonly use /opt/devsecops-ai. Docker and Ansible configurations need repair before they can be treated as verified installation routes.

Before you begin

Requirement	Why it matters
Git and access to the repository	Retrieve the source
WSL2 Ubuntu with Python 3.11 and venv support	Run the pinned backend dependencies
Ollama installed in the environment hosting inference	Serve the models
Free disk space for source, dependencies, and models	Model files are substantial; inspect your chosen downloads
Adequate RAM and optional supported GPU	Context length and model size determine resource demand
Network access during setup	Download dependencies, models, and initial embedding assets

The repository’s 4 GB GPU results are observations from one machine, not minimum requirements or throughput guarantees. See model reference before choosing models. Installation duration depends on download speed and hardware.

Confirm the terminal

If WSL is not installed, open Windows PowerShell as administrator, run wsl --install, restart Windows when requested, and complete Ubuntu’s first-launch user setup. If WSL already exists but Ubuntu does not, use wsl --list --online to find a distribution and wsl --install -d Ubuntu to install it. Follow Microsoft’s WSL installation instructions if your Windows version needs additional steps.

Use the exact distribution name shown by wsl --list --verbose in the commands below; it may include a version suffix.

In Windows PowerShell:

wsl --list --verbose
wsl -d Ubuntu

Inside WSL Ubuntu:

git --version
python3.11 --version
ollama --version
df -h .
free -h

If a command is unavailable, install that prerequisite before proceeding. Use the official WSL installation guide and Ollama documentation for host-specific setup.

Systemd for an existing service deployment

The terminal-based lab does not require a backend systemd service. For the service-management instructions, check PID 1 inside WSL with ps -p 1 -o comm=. If it is not systemd, edit /etc/wsl.conf with sudo nano /etc/wsl.conf and add or update the following section, preserving other settings:

[boot]
systemd=true

Save work in every WSL session, then run wsl --shutdown from Windows PowerShell; this stops all running distributions. Reopen your distribution and repeat the PID 1 check. See Microsoft’s systemd instructions for WSL version requirements. Enabling systemd does not itself install the Academy service.

Next: Install the local lab.