That's an interesting question, or at least it might be in the future, as it cuts right to the heart of what it means to be alive.
Right now no, an Arduino can't be sentient because it's not alive. If at some future point an Arduino could tick all the "is alive" boxes [0]:
- homeostasis
- organisation
- metabolism
- growth
- adaptation
- response
- reproduction
Then yes, there would be a good case for the Arduino being sentient.
However if you take a look at the list you'll see response is listed, that's because anything that is alive is sentient by definition.
Sentient by definition is really the point, sentient and sapient were defined specifically to codify the uniqueness of humanity.
All these other this are sentient, but only us, humans, and sapient.
The definitions although seemingly scientific are simply a re-enforcement of the biblical idea that the world was created for us, and that we can essentially do as we please, you can see this is the test for sapience, which is, is the specimen a member of the homo sapiens species.
Just for completeness, and to also touch on why this might become an important question; you'll also see organisation on that list.
Organisation is usually defined as being made up of one or more cells, which suggests that an Arduino, or any machine, will always fail to pass that test.
Essentially humans have defined life in terms of itself. This makes sense, we only have a single reference and as such we can't even begin to imagine what other forms of life may exist.
However, as we create ever more complex nondeterministic machines, or if we ever get serious about finding extraterrestrial life, this definition is going to have to change to allow for other possibilities.
[0]https://en.m.wikipedia.org/wiki/Life