Careers

Day to day

• Select parts and draw schematics; review other people's schematics.
• Lay out boards, run DRC and DFM, order prototypes, write assembly notes.
• Bring boards up: power sequencing, rail checks, programming, peripheral tests.
• Debug field returns with a scope, logic analyzer and thermal camera; write up root causes.

Must know (lessons)

• Circuit fundamentals and dividers
• Capacitor dielectrics, ESR and derating
• LDO vs buck, op-amp rules
• Stack-ups, vias, trace width
• Return paths, decoupling, EMI
• DFM and the fab process
• Systematic bring-up and debugging
• MOSFETs as switches

Tools

• KiCad or Altium
• LTspice or ngspice
• Oscilloscope, DMM, bench supply, logic analyzer
• Python for test scripts
• Git for design files

The interview loop

1. Phone screen: Ohm's law, dividers under load, why decoupling, LDO vs buck.
2. Take-home or live: review a schematic and find the three planted errors.
3. Onsite: design a 5 V to 3.3 V supply for a stated load, size a trace, explain a layout for a switching regulator.
4. Debug scenario: a board draws too much current at power-on; walk through your method.

Portfolio

• The KiCad motor-driver board from lesson 18, with Gerbers, a BOM and a one-page bring-up log.
• A schematic-review write-up of a public reference design with the issues you found.

Resume bullets to adapt

• Designed a 2-layer ESP32 motor-driver board in KiCad (TB6612FNG, 3.3 V LDO, reverse-polarity protection), passed fab DFM on the first submission and brought it up with a documented checklist.
• Diagnosed a brown-out reset caused by motor inrush on a shared ground return; fixed it with a 470 µF bulk capacitor and a star ground, verified on an oscilloscope.

Day to day

• Write drivers for I2C, SPI and UART peripherals from datasheets.
• Structure firmware around timers, interrupts and (sometimes) an RTOS.
• Profile and fix timing problems; keep control loops deterministic.
• Debug with serial logs, a logic analyzer and a debugger; ship OTA updates.

Must know (lessons)

• GPIO, ADC, PWM, I2C, SPI, UART, interrupts, RTOS
• Sensors and filtering
• Motor drivers and PWM
• Discrete PID with anti-windup
• Board bring-up from the firmware side

Tools

• C and C++
• Arduino, ESP-IDF, STM32Cube or Zephyr
• FreeRTOS
• Logic analyzer, JTAG/SWD debugger
• Python for tooling
• Git

The interview loop

1. Coding: bit manipulation, a ring buffer, a debounced button, a state machine.
2. Peripherals: I2C vs SPI, what pull-ups do, how a timer makes PWM, what an interrupt must never do.
3. System design: a sensor node that must run a year on a battery, or a 1 kHz control loop.
4. Debugging story: a bug you found with a logic analyzer.

Portfolio

• The robot firmware from lesson 26: fixed-rate control loop, encoder interrupts, serial telemetry, with a README.
• A PID write-up with step-response plots from the lab, explaining each tuning decision.

Resume bullets to adapt

• Implemented a 100 Hz control loop on an ESP32 with quadrature encoder interrupts, an ultrasonic driver and a PID speed controller, achieving under 10% overshoot and 1 s settling.
• Recovered a locked I2C bus in firmware with a clock-out routine and added 4.7 kΩ pull-ups, eliminating intermittent IMU drop-outs.

Day to day

• Write ROS 2 nodes in Python or C++ and wire them into launch files.
• Implement and tune controllers, odometry and navigation behaviors.
• Build and run simulations; measure the sim-to-real gap.
• Run field tests, log everything, triage failures.

Must know (lessons)

• Differential-drive kinematics and odometry
• PID and closed-loop control
• Behaviors, FSMs, planning
• Simulation and the sim-to-real gap
• ROS 2 concepts and QoS
• Sensor physics and fusion

Tools

• Python and C++
• ROS 2, Nav2, TF, RViz, rosbag
• Gazebo, Webots or Isaac Sim
• NumPy, OpenCV basics
• Linux, Docker, Git

The interview loop

1. Coding: implement odometry from encoder ticks, a PID class, or A* on a grid.
2. Math: rotate a vector between frames, explain a singularity, derive differential-drive velocities.
3. System design: a robot that follows a line and avoids obstacles; which nodes, which topics, which rates.
4. Behavioral: a time a simulated controller failed on real hardware.

Portfolio

• The robot plus a video of it doing obstacle avoidance and line following, and the controllers you wrote in the simulator lab.
• A ROS 2 node graph for the robot with a short design note on rates and QoS.

Resume bullets to adapt

• Built a differential-drive robot (ESP32, encoder gearmotors, ultrasonic and IR sensors) and validated obstacle-avoidance and line-following controllers in a browser simulator before deploying to hardware.
• Calibrated wheel diameter and track width with UMBmark, cutting odometry drift over a 10 m lap from 1 m to 12 cm.

Day to day

• Answer customer questions about datasheets, operating regions and layout.
• Reproduce field failures on an evaluation board; drive failure analysis.
• Write application notes and reference designs; present to customers.
• Characterize parts in the lab; analyze test and yield data.

Must know (lessons)

• Bands, doping, carriers
• Junctions and diode families
• MOSFET regions, Rds(on), gate charge
• Packaging, θJA and reliability
• Si vs SiC vs GaN, switching losses
• The value chain and the roles
• How wafers are made

Tools

• Datasheets and SPICE models
• Oscilloscope, curve tracer, thermal chamber
• Python or MATLAB for data
• PowerPoint, honestly
• Lab notebooks and bug trackers

The interview loop

1. Device basics: explain a MOSFET's three regions, why Rds(on) rises with temperature, why a body diode exists.
2. Application: a customer's MOSFET runs hot though it is on; what do you ask, what do you measure.
3. Failure analysis: a part fails at high temperature in the field; lay out the FA flow.
4. Communication: explain SiC to a non-engineer in two minutes.

Portfolio

• Three one-page device explainers (MOSFET switching, SiC vs Si losses, LDO vs buck) with your own diagrams.
• A failure-analysis write-up of a real or simulated board problem, structured as symptoms, hypotheses, measurements, root cause, corrective action.

Resume bullets to adapt

• Analyzed conduction and switching losses for a 100 A, 10 kHz SiC MOSFET stage (10 mΩ, 1 mJ per cycle) and recommended a gate-resistor change that cut EMI emissions while holding efficiency above 98%.
• Wrote an application note on logic-level MOSFET selection for 3.3 V microcontrollers, with measured Rds(on) at 2.5 V and 4.5 V gate drive.

Day to day

• Plan and run dyno, pack and charger tests; analyze the data in Python or MATLAB.
• Validate BMS limits, thermal behavior and charging curves across temperatures.
• Work interoperability issues between vehicles and chargers (CCS, NACS, ISO 15118, OCPP).
• Design and commission charging sites: transformer sizing, load management, layouts.

Must know (lessons)

• The powertrain block diagram, 400 V vs 800 V, regen
• Induction vs PM motors, field weakening
• Inverters, PWM and FOC
• Cells, packs, BMS, charging curves
• The Tesla lineage and motor advances
• AC and DC charging, connectors, OCPP, site design
• Si vs SiC in traction inverters

Tools

• Python, pandas, MATLAB
• CAN tools (CANalyzer, SavvyCAN), OCPP simulators
• HV-safe multimeters, insulation testers, dynos, thermal chambers
• AutoCAD or site-design tools for EVSE
• ISO 15118 and SAE J1772/J3400 references

The interview loop

1. Fundamentals: how three phase currents make a rotating field, why Tesla moved to PM motors, why charging tapers past 80%.
2. Numbers: cable current at 150 kW on 400 V vs 800 V, pack kWh from a 96s46p layout.
3. Test planning: validate a BMS charge-current limit across −10 °C to 45 °C.
4. Charging: design load management for eight 150 kW dispensers on a 600 kW service.

Portfolio

• An EV charging site design you have done, written up as requirements, loads, transformer, layout, compliance.
• A motor-and-inverter explainer with the three-phase and FOC visualizations from this site, in your own words.
• A charging-session analysis: SoC, power and temperature over time, with the taper explained.

Resume bullets to adapt

• Designed EV charging sites with load-managed DC fast chargers, sizing transformers and switchgear for shared-power dispensers and ADA-compliant stall layouts.
• Modeled a 96s46p 2170 pack (76.3 kWh nominal) and its CC-CV charging taper versus temperature to set realistic charge-time expectations for site planning.