Hi everyone,
I've been working with Ghost Solution Suite mainly for traditional OS deployment and imaging, but recently started wondering how well it fits into more modern workflows - especially those involving AI-ready environments.
For example, in scenarios where machines need to be pre-configured not just with OS and drivers, but also with:
AI frameworks (like Python environments, PyTorch, etc.)
optimized dependencies for inference workloads
preloaded datasets or model checkpoints
it feels like deployment tools might need to evolve a bit beyond classic imaging.
I've been reading more about how current AI systems are structured - particularly how environments are prepared for consistent deployment and execution - and came across a concise overview here: https://techstoffe.com/
It got me thinking about whether tools like Ghost can still play a role in these newer pipelines, or if teams are moving toward more container-based or hybrid approaches instead.
So I wanted to ask:
Has anyone used Ghost Solution Suite for deploying AI/ML-ready systems?
Are there best practices for maintaining reproducibility in such setups?
Or is it more practical to combine Ghost with other tools (e.g., containers, config management)?
Would be great to hear how others are approaching this.
-------------------------------------------