CAD scripting Expert

Headquarters:

Overview

We are building evaluation benchmarks that test whether AI agents can complete real professional work from start to finish. We are hiring expert mechanical design and CAD automation engineers to turn real, paid CAD tasks from their own work into expert-level Terminal-Bench challenges. Each task includes the files and systems it runs on, a working solution and a check that proves the result is correct.

***THIS IS A TASK BASED COMPENSATION MODEL. PLEASE CHECK THE APPLICATION QUESTIONS FOR MORE DETAILS.***

Responsibilities

• Design expert-level Terminal-Bench tasks by turning real, paid mechanical CAD work from your own experience into challenges an AI agent must solve end to end.

• Define each task precisely, including the files, systems and environment it runs on, so an AI agent can attempt it without ambiguity or hidden assumptions.

• Write a working reference solution using scripted CAD, such as FreeCAD or CadQuery Python, that rebuilds parametric parts correctly after parameters change.

• Build automated checks and tests that prove the result is objectively correct, covering geometry, dimensions, tolerances and file outputs such as STEP or IGES.

• Calibrate difficulty so each task takes an expert about 60-75 minutes to solve, and explain how a real-world mistake would be caught by the checks.

• Iterate on tasks based on reviewer feedback, documenting the design intent, tools, data and rules used so others can reproduce and validate your work.

Skills/Experience

• 3+ years designing real mechanical parts that were manufactured or shipped, doing the hands-on work yourself (not only studying, selling, managing or teaching it)

• Hands-on CAD work within the last 3 years, with the ability to name the systems, tools, data and rules you used

• Builds or modifies designs through scripts (FreeCAD or CadQuery Python, OpenSCAD, Onshape FeatureScript), not only through GUI work

• Reads engineering drawings, tolerances and GD&T, and can explain how a design mistake would be caught

• Technical fluency with Python or another language and the command line, plus files such as CSV/JSON, APIs or databases

• Availability to start around early November 2026 and take on tasks on a flexible, task-based schedule

Nice-to-have

• Parametric design-intent experience

• Open-source hardware contributions or projects

• Docker and Git experience

• Experience writing automated tests

• Experience with SolidWorks, Fusion or Onshape APIs and STEP/IGES workflows

Process

• Task-based vetting: you complete a vetting task instead of an interview

• Paid per task, with each task taking about 60-75 minutes to solve

• Flexible, task-based work cadence, with an expected start around early November 2026

To apply: https://weworkremotely.com/remote-jobs/toptal-cad-scripting-expert