From the Hardware & Embedded course. Two files: the electronics/embedded half, and the RF/IoT half.

Files

  1. Electronics & Embedded — resistor sizing, decoupling, pull-ups, UART/I2C/SPI, interrupts, regulators, level shifting, ADC, PWM, board bring-up, RTOS
  2. RF & IoT — radio selection, link budget, antennas, MQTT, battery-life estimation, edge vs cloud, offline devices, PoE, safe OTA updates

What these interviews actually test

Three things, in descending order of how much they separate candidates:

1. Can you do the arithmetic out loud? Hardware interviews are unusually numerical. Size a resistor, compute a regulator’s dissipation, estimate battery life, do a link budget in dBm. These are all one-line calculations, and being fluent at them reads as experience because the people who’ve built things do them constantly. The people who haven’t reach for a definition instead.

2. Do you know why a component is there? “A decoupling capacitor smooths the supply” is a definition. “It’s a local charge reservoir because trace inductance can’t deliver a nanosecond current spike, which is why placement matters more than value” is understanding. Every 🔥 question here has that split.

3. Do you debug by bisection or by guessing? Q12 — “the board doesn’t power on” — is the hardware version of “the service is down.” Same method, different instrument. Interviewers care far more about the method than about any component.

The thing that beats every answer here

Cite your own board. The IoT Bridge PCB is a real design with real component choices and real tradeoffs written down. “Here’s why I picked that regulator, and here’s what I’d change in rev 2” outperforms any prepared answer in this folder — and it’s a story almost no candidate has.

↑ Back to Hardware & Embedded