AI for embedded IoT development is changing the way engineers solve technical problems.
This morning we integrated a BLE pressure sensor with a Senquip device. There was no documented packet format, no existing code, and no prior integration to work from.
Working with Claude, we decoded the sensor’s BLE advertisement packet from scratch.
By applying known pressures—including grabbing a compressor to generate a precise 6 bar—and observing which bytes changed, we identified the pressure field, confirmed the units, and developed a working integration script in under an hour.
The result wasn’t just a proof of concept. It was clean, readable production-quality code that correctly:
- Decoded BLE GAP advertisement packets
- Parsed binary data using
SQ.parse() - Handled the correct byte order (endianness)
- Identified and corrected byte-offset errors during development
I’ve been cautious about relying too heavily on AI for engineering work. However, the more I use it, the more I appreciate the productivity gains.
In this case, Claude demonstrated a detailed understanding of the Senquip scripting environment, the BLE GAP API, and the associated parsing functions. It wasn’t simply generating code—it was helping reason through an unfamiliar protocol, validating assumptions, and refining the implementation until the decoded values matched real-world measurements.
The engineering didn’t disappear.
Instead, AI removed much of the repetitive investigation, allowing us to focus on testing, validating results, and understanding the system.
If you’ve been hesitant to introduce AI into your engineering workflow, it may be worth taking another look.
The biggest advantage isn’t that AI writes code.
It’s that it dramatically reduces the time between identifying a problem and having a working solution.
