$60–150/hr · micro1
Create high‑fidelity 3‑D simulation assets and ensure they behave like real hardware for AI training projects.
What you would do
- Model and texture objects using CAD or 3‑D tools
- Import and set up assets in a chosen simulation engine
- Calibrate physics properties to mirror real‑world behavior
- Collaborate with scientists to verify accuracy
- Debug and resolve fidelity gaps
Who they want
- Proven hands‑on work with at least one robotics/physics simulator
- Strong background in 3‑D modeling or CAD
- Understanding of rigid‑body dynamics and material physics
- Excellent communication for interdisciplinary teams
- Remote‑first availability, flexible hours
Main skills
What the interview asks about
1.Asset geometry translation
The interviewer wants to confirm you can faithfully reproduce real‑world dimensions, which is critical for training AI on accurate inputs.
For example: “Describe a situation where you had to model a complex laboratory instrument for Gazebo, ensuring the scale was within 2 mm of the physical part, and how you verified that accuracy.”
2.Physics parameter calibration
Accurate mass and friction values directly affect simulation outcomes, so they’ll assess your systematic approach to tuning them.
For example: “Tell me about a time you adjusted the friction coefficient of a conveyor belt in Isaac Sim to match observed slip rates, including the data you used and the iteration count.”
3.Cross‑tool integration
Switching between platforms can introduce inconsistencies; they’ll test your ability to maintain fidelity across tools.
For example: “Explain how you migrated a robot arm model from Unity to MuJoCo while preserving joint limits and sensor placements, and what challenges arose.”
4.Stakeholder validation
Collaboration with scientists ensures the model meets research goals, so they’ll explore your communication and validation workflow.
For example: “Give an example of a review session with a biomechanics researcher where you had to reconcile differing opinions on material stiffness, and how you reached consensus.”
5.Troubleshooting simulation drift
Detecting and fixing divergence between simulated and real behavior is essential for reliable AI training, so they’ll probe your diagnostic process.
For example: “Recall a case where a simulated drone’s flight path deviated from real test data; detail the steps you took to identify the root cause and correct it.”
A task you may get
You may be asked to import a CAD file of a handheld tool into Isaac Sim, assign realistic material properties, and demonstrate that its motion matches a provided video clip within a 5 % error margin.
How to prepare
- Gather a portfolio of simulation projects with before‑and‑after fidelity metrics
- Refresh your knowledge of physics parameters (mass, inertia, friction) for at least two platforms
- Practice explaining your validation process to non‑engineers
- Set up a quick demo of importing a model and tweaking dynamics in a sandbox environment
The facts
- Pay
- $60–150/hr
- Open to
- Bangladesh, Hong Kong, India, Indonesia, Japan, Kazakhstan, Kyrgyzstan, Malaysia, Pakistan, Philippines, Singapore, Sri Lanka, Taiwan, Thailand, Uzbekistan, Vietnam, Austria, Belarus, Belgium, Denmark, France, Germany, Greece, Italy, Netherlands, Portugal, Russia, Spain, Switzerland, United Kingdom, Argentina, Brazil, Chile, Colombia, Mexico, Peru, Algeria, Bahrain, Egypt, Iraq, Jordan, Kuwait, Lebanon, Libya, Morocco, Oman, Palestine, Qatar, Saudi Arabia, Tunisia, United Arab Emirates, United States, Canada, Nigeria, Kenya, South Africa, Ghana, Ethiopia
- Field
- Applied Engineering
- Role type
- Specialist
- Posted
- 10/3/2026
- Places left
- 5
We wrote this page from the public micro1 listing. It may be out of date, so read the full posting before you apply.