Embedded & IoT
Edge AI & TinyML
6 practice questions. Free questions open a full answer guide; the rest unlock with Pro.
- To run a small neural network on a microcontroller, models are usually converted from 32-bit floats to 8-bit integers. What does that quantization buy you, and what's the catch?
- You've quantized a model to run on-device and it passes on your bench, but a fleet in the field starts giving worse predictions over time. How do you diagnose and address that?Go Pro
- Your team trained a model that works well, but it won't fit the flash, RAM, and latency budget of the target microcontroller. How do you get it to run on-device?Go Pro
- You've quantized and deployed a TinyML model to the device, but its predictions are noticeably worse than they were in your training environment. How do you debug the gap between the model you trained and the one running on the hardware?Go Pro
- Product wants a small machine-learning model — say keyword spotting or anomaly detection — running directly on a battery-powered microcontroller instead of in the cloud. How do you decide if that's feasible and get a model to actually fit on the device?Go Pro
- When you have a sensor that produces a lot of data, how do you decide whether to run a model on the device itself versus sending the raw data to the cloud to process?Go Pro
Want questions matched to your role? Paste a job title, job description, or CV for a personalized set, or go Pro to unlock the full bank.