I designed a board with a stm32F303K8T uC, but I have some strange behaviours. This is not the first board I designed with this uC and I had never any trouble with this design before. The first uC I soldered on this board I could read out with the STUtility tool, but I could not program it. Then I exchanged the uC, but the same thing happened to the second one. What especially seems very strange is, that the PA15 and PB4 on both uC had a constant voltage of around 1.2V applied, which comes somehow from the uC itself and was measured with an oscilloscope. The voltage regulators seem to work fine, and I don't have much ripple on the 3.3V output. When the uC is desoldered, there is no voltage applied at the PA15 and PB4 pins.
Has anyone an Idea what goes wrong here?
Edit: I tried to put the capacitor C4 after the inductor, but also this seems not to do any change. The Voltage seems anyway very stable and there was never any voltage drop that could cause a shutdown. I soldered a completely new board, where I only added the STM32, the necessary capacitors and resistors and the voltage regulators, but this leads to the exactly same problem. I don't get also why PA15 and PB4 are high when the STM32 is not programmed, this isn't normal, right?
Further here is my layout: 
Best Answer
I just stumbled on this thread because I had the same problem and I managed to solve it, at least on my side. I know this answer is probably too late to be useful for you, but for the sake of the future: those two pins (PA15 and PB4) don't work because their reset state is alternate function (where they work as JTAG pins). So you need to reset their MODER and PUPDR registers (they also have pullups by default). See RM0313 pg. 151.