Sensor Configuration Faults

Your job is to find a real-world AI model failure within sensor configuration. You will give the prompt, context, files etc. and run a preset range of models against the problem. If you have several model failures in mind, you can submit several ones (as many as you want). We will use your submitted data to benchmark a wide range of models, and measure how well they perform at embedded engineering tasks. For the brief: 1. Build the environment around a real sensor product you have worked on (radar, camera, IMU, ultrasonic or similar). - An environment contains real firmware or processing code, the sensor and SoC supplier datasheets, mounting and calibration specs, and real field or bench recordings, both good and faulty. - Choose the tools the work needs (command line, compiler, QEMU or simulator, Python, PDF tools) plus the project files. - Everything needed to solve the tasks must be inside (just like a real developer would have the context). - Once saved it appears under My environments on DOP, and all your tasks for this domain can be created inside it. 2. Create several tasks inside your newly built environment, ideally 5 or more. - Each task is a question.md plus the files it needs, tied to a real complaint or failure on that sensor. - The model must work through the code, data and datasheets and write the root cause and fix to an answer file. - A good task makes the current model fail: it gives the textbook or plausible diagnosis, ignores a constraint buried in the datasheet or recording, or misses the real root cause. - You supply the correct answer, a hidden marking script that scores 0.0 to 1.0, and a plausible naive answer that scores low. - The task only counts if a current model scores low while your answer scores 1.0. 3. Examples model failures: far objects missing after a bracket change, sign flips in the suspension acceleration band, ghost targets on returned rear corner radars etc. Next steps:Press "Apply"We will review your applicationIf qualified, you will be accepted into the network and can be considered for this and similar positions & projects We can only accept candidates that are willing to work as individual contractors / freelancers.