Once Lectra had a real shell, real git, and real Python running on the iPad, the next question was what you build on top of them. Two things landed. The first is a coding agent: you describe what you want, and it reads your files, proposes edits, and runs your project's tests, working through the task a step at a time, showing every file change and command for your approval before anything happens, and refusing anything risky.
It follows a deliberate rhythm instead of free-forming: write down the spec, break it into a checklist, do one item at a time, review the result, and only then commit, the same disciplined loop a careful engineer would use, kept in plain files you can read and edit yourself. It never pushes to GitHub unless you ask.
The second addition is real SSH. The built-in terminal can now connect to another machine, a lab server or your own computer, and behaves like a genuine terminal, colors, full-screen programs and all, so the tools you'd run at a desk run from the iPad too. Between the agent and SSH, the iPad stops being a place you only read code and becomes a place you can actually work on it.