Yes! That’s one of the key advantages here. The agent works from actual N|Solid runtime evidence (CPU profiles, heap snapshots, telemetry, and more) so you can validate its recommendations against the behavior of the running application, not just the source code.
And if you have a deeper technical question or a specific use case you’re trying to solve, I’d be happy to connect you with someone from our team who can dig into it with you in more detail 💚☺️