RFIC Design Interview Guide (2026)

RFIC interviews test a specific kind of intuition: the ability to think at gigahertz frequencies where wires become transmission lines, impedance matching is everything, and noise is measured in decibels referenced to 50 Ω. This guide covers the questions that come up most for LNA, mixer, VCO, and transceiver roles at companies like Qualcomm, Broadcom, Apple, MediaTek, and Skyworks.

It builds on general analog fundamentals (see the top 50 analog questions) and pairs with our AMS vs RFIC comparison if you're still choosing a path.

How do matching and the Smith chart come up in RFIC interviews?

1. Why do we match to 50 Ω? It's the standard system impedance balancing power handling and loss for coax/PCB; matching maximizes power transfer and minimizes reflections.

2. What is the Smith chart and how do you use it? A normalized impedance/admittance plane. Series elements move you along resistance/reactance circles; shunt elements move along conductance/susceptance circles. You navigate from a device impedance to the center (matched) by adding L/C.

3. Series inductor vs shunt capacitor: which way on the chart? A series L moves clockwise along constant-resistance circles; a shunt C moves clockwise along constant-conductance circles. Knowing the directions cold is expected.

4. What is return loss and VSWR? Return loss (dB) quantifies reflected power; VSWR is the standing-wave ratio. A good match is high return loss / VSWR near 1.

S-parameters

5. Define S11, S21, S12, S22. Input reflection, forward gain, reverse isolation, output reflection. They describe a network in terms of incident/reflected waves at each port.

6. Why use S-parameters at RF instead of Z/Y? At high frequency, opens/shorts are hard to realize for Z/Y measurement; S-parameters use matched terminations, which are practical.

7. What is the difference between the power gain definitions? Transducer, available, and operating power gain differ by what's assumed matched at input/output. Be ready to distinguish them.

What questions come up about low-noise amplifiers (LNAs)?

8. What is noise figure? The degradation in SNR from input to output, in dB; NF = SNR_in/SNR_out. The LNA's NF dominates the receiver chain (Friis).

9. State Friis' formula and its implication. F_total = F1 + (F2−1)/G1 + .... The first stage's noise and gain dominate, which is why the LNA must be low-noise and have gain.

10. Why inductive source degeneration in an LNA? It creates a real input impedance for matching without adding the thermal noise of a physical resistor, enabling simultaneous noise and power match.

11. Noise match vs power match: what's the tension? The impedance for minimum noise figure (Γopt) generally differs from the conjugate power match. LNA design reconciles them; degeneration helps bring them together.

12. What sets LNA linearity? Device size, bias, and degeneration; quantified by IIP3 and P1dB.

Mixers

13. What does a mixer do? Frequency translation by multiplying RF and LO signals, producing sum and difference frequencies.

14. Active vs passive mixer trade-offs? Active (Gilbert cell) gives conversion gain but more noise/power; passive gives better linearity and lower 1/f noise but loss.

15. What is conversion gain? Ratio of IF output to RF input signal level.

16. What problems are LO leakage and self-mixing? LO feedthrough to RF/IF and DC offset in direct-conversion receivers from LO self-mixing, a key zero-IF challenge.

17. What is image rejection? Suppressing the unwanted image frequency; addressed with image-reject (Hartley/Weaver) architectures or filtering.

Oscillators and VCOs

18. What is phase noise and why does it matter? Short-term frequency instability skirting the carrier; it corrupts adjacent channels and limits modulation accuracy.

19. State Leeson's model qualitatively. Phase noise improves with higher tank Q, higher signal power, and lower device noise; it worsens close to the carrier (1/f³, 1/f² regions).

20. Why does tank Q matter so much? Higher Q means lower phase noise and lower power for a given performance. On-chip inductor Q is often the bottleneck.

21. LC vs ring oscillator trade-offs? LC: low phase noise, narrow tuning, area-hungry inductors. Ring: wide tuning, compact, but much worse phase noise.

22. What sets VCO tuning range? Varactor capacitance ratio and switched-capacitor banks; there's a tension between tuning range and phase noise (Kvco sensitivity).

Linearity and system metrics

23. What is P1dB? The input/output power where gain compresses by 1 dB, a large-signal linearity measure.

24. What is IP3 / IIP3? The extrapolated point where third-order intermodulation products equal the fundamental; predicts in-band distortion from nearby interferers.

25. Why does linearity matter in a receiver? Strong adjacent signals create intermodulation that falls in-band, desensitizing the receiver. The whole chain budgets linearity.

Passives and layout

26. Why is on-chip inductor Q limited? Substrate loss and metal resistance; patterned ground shields and thick top metal help.

27. When do wires become transmission lines? When physical length is a non-negligible fraction of the wavelength; then you must model characteristic impedance and delay.

28. Why EM simulation at RF? Lumped models miss coupling and distributed effects; EM tools (momentum/HFSS-style) capture real passive behavior.

How should you prepare for an RFIC interview?

RFIC interviews reward engineers who can reach for the Smith chart and reason about noise figure and linearity in the same breath. Tips:

• Practice matching problems on the chart until the moves are automatic.
• Internalize Friis: it explains why the LNA matters most.
• Always separate noise, gain, and linearity when analyzing a stage, then discuss the trade-off.
• Know your metrics' units: NF in dB, IIP3 in dBm, phase noise in dBc/Hz at an offset.

The best way to pressure-test this is a mock interview with an engineer who designs radios for a living. On MockVise you can book a mock interview with verified RFIC engineers and get a candid debrief on your matching, noise, and linearity reasoning, exactly where real RF loops separate candidates.

Think in frequency and power, respect the 50 Ω world, and let the Smith chart be your friend.

FAQ

Q: Which roles and companies does RFIC interview prep cover? LNA, mixer, VCO, and transceiver roles at companies like Qualcomm, Broadcom, Apple, MediaTek, and Skyworks.

Q: Why is Friis' formula so central in RF interviews? The first stage's noise and gain dominate the receiver chain. That is why the LNA must be low-noise and also have gain.

Q: What is the tension between noise match and power match? The impedance for minimum noise figure generally differs from the conjugate power match. LNA design reconciles the two, and inductive source degeneration helps bring them together.

Q: Why does tank Q matter so much for a VCO? Higher Q means lower phase noise and lower power for a given performance. On-chip inductor Q is often the bottleneck.