Jax, the agent

Jax is Nexma’s spatial agent. You describe what you want; Jax does the work and shows you each step. The interface is natural language. The output is real, auditable project changes.

What Jax can do

• Generative design. “Plan a fiber distribution network for every household in

this polygon.” Jax composes layout, validates against the rules of your domain, and writes the result to your project.

• Spatial Q&A. “How many active alerts are within 5 km of this asset, broken

down by severity?” Jax searches your project, runs the geometry, returns a chart or list.

• Intent-driven edits. “Move every cabinet within 50 m of the floodplain to

higher ground; keep the design valid.” Jax reads the constraint, plans the moves, runs them.

• Explanations. “Why did the optimizer pick this route?” Jax replays the

reasoning in plain language, grounded in the actual run.

How to talk to Jax well

A good prompt has three pieces:

• Objective — what you want produced.
• Scope — where it should apply (a polygon, a layer subset, a time window).
• Constraints that matter to you — anything beyond what your project already

enforces.

Don’t restate rules Jax already knows. If your project says splitter ratios max at 1:32, you don’t need to tell Jax. Be specific about anything non-default.

What Jax is not

• Not a chatbot. Jax doesn’t answer trivia. Ask for the weather and it will

offer to add a weather feed to the project instead.

• Not a black box. Every action is visible on the timeline. You can step backwards.
• Not magic. Jax respects the rules and data you give it. Get those right and you

get great answers; get them wrong and you get a confidently wrong answer that solves the wrong problem.

When Jax pushes back

Sometimes Jax will tell you a request is infeasible — the polygon has no road access; the constraint set is over-specified; the math didn’t converge. Take that seriously. Loosen a constraint, narrow the scope, or ask Jax for a counter-proposal.