ESP32-C3 damaged through production (ESD?)

[TH9185]

Hallo

In ower new Project we use the ESP32-C3 on a small PCB to sample a I2C sensor.
Now on our first Prototype series we observed quite a big number of broken devices (12/100pcs.). We suspect ESD but we are not sure. The PCBs were professionally build unter ESD save surrounding. The programming and handling of them until build in to the enclosure happened on buiro surrounding.
Do you have any idea what could be the Problem or if it is ESD caused how could we improve ESD resistance.
The schematic is attached.
Thanks in advanced we are grateful for all help.

[ESP_Sprite]

In what way is the device 'broken'? Before you decide it's 'ESD' (something that can't really be disproven unless you have real fancy equipment), have you already ruled out other things like bad soldering, mechanical issues, out-of-tolerance components etc?

EDIT: One issue I see is that that LDO can only deliver 150mA. Per the hardware design guidelines: 'If you use a single power supply, the recommended voltage of the power supply for ESP32-C3 family is 3.3 V, and its recommended output current is 500 mA or more.'

[TH9185]

Hay thanks for the response.

The CPU is completely dead. No response on serial output and no funktion of GPIO's. When I solder a new ESP32 it works on same PCB.

Here i listed the failure rate with timing.
4.7.2023 While initial flashing, 7/100 broke (bare PCB)
24.7.2023 While EOL testing for first project, 3/46 broken (after complete assembly)
3.8.2023 While EOL testing for second project, 3/8 broken (after complete assembly)
8.8.2023 On shipment to first project 1/9 tested was broken, others will be tested while installation.
9.8.2023 1/3 broke in development team with cable but without enclosure.
11.8.2023 Included fixed ones, 23 left in production (bare PCB untested)

I suspect ESD because it looks so random. They are breaking mostly while they are not connected to any system.
We have some running on pilot tests and as soon they are build in they seem to work without problem. (only small scale tests 5 units)

I will have a look on the Power supply now with oscilloscope but as long the BLE functionality is not used the current required should be less than 50mA so the DCDC should be fin in our application.