by A quality engineer keeps a manufacturing process producing parts that meet print, spec, and customer requirements, and builds the paperwork trail that proves it. The job mixes statistics, root cause investigation, and a lot of writing: control plans, FMEAs, PPAP packages, and corrective action reports.
Interviewers are testing 2 things at once. Can you actually run the tools, SPC, FMEA, MSA, 8D, without a textbook in front of you? And will you hold the line on a quality hold when a plant manager wants the line running again?
Expect a mix of technical questions on specific methods, and behavioral or situational questions about the moment a nonconformance shows up and everyone wants a fast answer instead of the right one.
Quality engineer at a glance
| Item | Details |
|---|---|
| Closest BLS occupation | Industrial engineers (quality engineer is a listed job title under this SOC code, per O*NET) |
| Median pay | $102,440 per year (BLS, May 2025) |
| Job outlook | 12% growth from 2025-2035, about 23,100 openings a year (BLS) |
| Typical employers | Automotive and aerospace suppliers, medical device makers, electronics and machining shops |
| Education | Bachelor's degree in engineering, industrial technology, or a related field |
| Certification | ASQ Certified Quality Engineer (CQE): 8 years of work in the CQE body of knowledge, with a bachelor's degree waiving 4 years |
| Key tools | Minitab or JMP, SPC and gauge software, CAD viewers, PPAP and FMEA templates, a CAPA or nonconformance database |
| Interview format | Phone screen, then a technical interview or take-home problem, then an on-site panel with a plant tour |
How the interview usually works
Every plant runs its own process, and the steps below describe what's common rather than a fixed sequence.
- Recruiter or HR screen. Confirms your industry background, shift availability, and whether you have or are working toward a CQE or Six Sigma belt.
- Technical interview. A quality manager or engineering lead asks you to walk through a control chart, an FMEA, or a past root cause investigation, often with a whiteboard or a sample dataset.
- Case or problem-solving exercise. Some employers hand you a nonconformance report or a set of measurement data and ask you to find the root cause on the spot.
- On-site panel and plant tour. You'll likely meet production supervisors and an engineering manager, and walk the floor to see the process you'd support.
- Offer and background check. Manufacturing roles commonly include a background check and sometimes a basic physical or drug screen before a start date.
General and background questions
1. Why do you want to work in quality engineering?
Why they ask: They want to know if you're drawn to the investigative and statistical parts of the job, not just the title.
How to answer: Point to a specific moment that pulled you toward root cause work or process control, not a general interest in manufacturing.
Sample answer: In my last co-op, I watched an engineer trace a weld porosity problem back to a shielding gas flow rate that had drifted 15% over 3 months, using nothing but SPC charts and patience. That's what got me. I like the part of the job where a defect looks random until you find the one variable that explains it. I've since run 2 root cause investigations of my own on a stamping line, and both times the answer was in the data, not in a guess.
2. What quality certifications do you have, and which are you working toward?
Why they ask: Certifications signal a baseline of shared vocabulary and method, especially for CQE and Six Sigma.
How to answer: State what you hold, what you're studying for, and how you use the material day to day, not just the credential name.
Sample answer: I'm a Six Sigma Green Belt through my last employer's internal program, and I'm scheduled to sit for the ASQ Certified Quality Engineer exam in March. I qualify with a bachelor's degree in mechanical engineering plus 4 years in a decision-making quality role, which covers the CQE's experience requirement. I've been studying the sampling and reliability sections hardest, since my daily work leans more on SPC and FMEA than on the statistics I haven't used since school.
3. Which industries and part types have you worked with as a quality engineer?
Why they ask: SPC and FMEA look different on a stamped bracket than on an injection-molded medical component. They want to know how much translation you'll need.
How to answer: Name the industry, part complexity, and any regulatory framework you worked under.
Sample answer: Three years in tier-2 automotive, mostly stamped and welded brackets under IATF 16949, then a year in a contract electronics shop building PCB assemblies for industrial controls. The biggest adjustment was inspection method: automotive was heavy on go/no-go gauges and CMM reports, electronics was heavier on visual inspection under magnification and functional test data. I expect this role to be closer to my automotive experience, given the PPAP requirement in your job posting.
Technical questions
4. How do you decide when a process is out of control?
Why they ask: This is the core judgment call in SPC. They want to know you use rules, not a gut feeling.
How to answer: Name the control chart rules you use, such as a point beyond 3 sigma or a run of points on one side of the centerline, and what you do next.
Sample answer: I watch for a point outside the 3-sigma control limits, 7 or more consecutive points on one side of the centerline, or a clear trend of 6 points rising or falling. Any of those triggers a stop-and-check, not an automatic line stop, since I still need to confirm it's a process shift and not a bad measurement. On a bore diameter chart last year, I caught a 6-point downward trend 2 hours before it would have hit the lower spec limit, which gave the operator time to adjust the tool offset instead of scrapping parts.
5. Walk us through how you'd set up an X-bar and R chart for a new characteristic.
Why they ask: This checks whether you understand subgroup sampling, not just the finished chart.
How to answer: Cover subgroup size, sampling frequency, how you calculate control limits from real data, and how long you'd run before locking the limits.
Sample answer: I'd start with subgroups of 5 consecutive parts, pulled every 30 minutes for a high-volume line, and collect at least 20 to 25 subgroups before calculating control limits, so the limits reflect real process variation and not a lucky morning. I'd plot the range chart first, since an out-of-control range chart makes the X-bar limits meaningless. Once both charts are stable, I lock the limits, note the process conditions on the control plan, and only recalculate them after a documented process change.
6. What's the difference between Cp, Cpk, and Ppk, and which do you report to a customer?
Why they ask: Confusing these 3 numbers is a common and costly mistake in a PPAP submission.
How to answer: Define each briefly and say which one is typically required at PPAP versus ongoing production.
Sample answer: Cp compares the spec width to process spread without accounting for centering, so a process can have a great Cp and still make bad parts if it's off-center. Cpk factors in centering, using short-term, within-subgroup variation. Ppk uses the overall long-term standard deviation across the whole study. For a PPAP submission I report Ppk from the initial production run, usually targeting 1.67 or better, since that reflects real process behavior over the study period rather than the best-case short-term number.
7. Explain the PPAP process and the 5 submission levels.
Why they ask: PPAP is the standard new part approval process across automotive and similar supply chains. They want to know you can run it, not just define it.
How to answer: Describe what PPAP proves, the 18 elements at a high level, and how the 5 levels differ.
Sample answer: PPAP proves that a supplier's process, run at production rate with production tooling, makes parts that meet the print and spec. It covers up to 18 elements: design records, process flow, control plan, FMEA, MSA, dimensional results, material certs, and the Part Submission Warrant, among others. Level 1 is a PSW only. Level 3, the default most customers require, adds product samples and complete supporting data. Level 5 keeps everything on-site for the customer to review in person. I've submitted mostly Level 3 packages, usually with a 300-piece capability study behind the Ppk numbers.
8. How does APQP fit with PPAP on a new part launch?
Why they ask: They want to see you understand the timeline, not just 2 acronyms.
How to answer: Explain APQP as the planning process across the program, with PPAP as the final approval gate.
Sample answer: APQP is the 5-phase planning process that runs from the design concept through launch: planning, design and development, process design, validation, and feedback. PPAP sits at the end of the process validation phase, as the formal proof that the process is ready. On my last program, APQP started 14 months before start of production, with the FMEA and control plan drafted during process design and finalized before the PPAP run at month 12. Skipping ahead to PPAP without finishing the APQP steps is how teams end up submitting a package that fails on the customer's first review.
9. Walk us through how you'd run an FMEA and calculate RPN.
Why they ask: FMEA quality varies a lot between engineers who treat it as a form and engineers who use it to actually prevent failures.
How to answer: Cover the team-based approach, the severity, occurrence, and detection scoring, and what you do with a high RPN.
Sample answer: I pull in the process owner, an operator, and sometimes a customer quality rep, since one person's FMEA misses failure modes. For each process step we list potential failure modes, effects, and causes, then score severity, occurrence, and detection from 1 to 10 and multiply for the RPN. On a recent bracket line, a weld crack failure mode scored severity 8, occurrence 4, detection 5, for an RPN of 160. That triggered an action: adding an in-process dye penetrant check, which dropped detection to 2 and the RPN to 64. I revisit the FMEA whenever a process change or a real nonconformance shows a failure mode we missed.
10. What's your process for root cause analysis on a nonconformance?
Why they ask: This is the single most common day-to-day task in the role, and a vague answer usually means shallow experience.
How to answer: Name a specific method, such as 5 Whys or a fishbone diagram, and describe how you confirm the root cause rather than just guessing at one.
Sample answer: I start by containing the nonconforming parts and sorting the affected lot, then build a fishbone diagram across the 6 M's: machine, method, material, manpower, measurement, environment. From there I run 5 Whys on the most likely branches, but I don't stop until I can reproduce or disprove the cause with data. On a recent flash defect in molded parts, "why" led to a mold temperature drop, which led to a failed heater cartridge that hadn't tripped an alarm. I confirmed it by swapping the cartridge and rerunning 50 parts clean before closing the investigation.
11. Explain the difference between a containment action and a corrective action.
Why they ask: Customers and auditors expect these to be documented separately, and mixing them up is a common CAPA writing mistake.
How to answer: Define containment as immediate and temporary, and corrective action as addressing the root cause so it doesn't recur.
Sample answer: Containment stops the bleeding: sort existing inventory, add a temporary 100% inspection, and notify the customer if shipped product might be affected. Corrective action fixes the root cause so the same defect doesn't come back, which might mean a poka-yoke fixture, a revised work instruction, or a process parameter change. On an 8D I ran last year, containment was a same-day sort of 3,400 parts in the warehouse. The corrective action, a light curtain that stopped the press if a part wasn't seated correctly, took 3 weeks to install and verify.
12. What's the difference between ISO 9001 and IATF 16949?
Why they ask: Many quality engineers move between industries, and this checks whether you understand what the automotive standard adds.
How to answer: State that IATF 16949 builds on ISO 9001:2015 with automotive-specific requirements.
Sample answer: ISO 9001:2015 is the general quality management standard, built around a process approach and continual improvement, and it applies to any industry. IATF 16949 uses ISO 9001 as its foundation and adds automotive-specific requirements: PPAP, control plans, a documented FMEA process, manufacturing feasibility reviews, and stricter requirements for contingency planning and supplier development. My last plant held both certifications, and the IATF audit always dug much deeper into our control plans and PPAP files than the ISO-only audit did.
13. How do you prepare for and conduct an internal quality audit?
Why they ask: Internal audits are how a plant catches problems before an external auditor or a customer does.
How to answer: Describe your prep, what you check against, and how you handle a finding.
Sample answer: I pull the process's control plan, work instructions, and the last audit's findings before I walk the floor, so I know what I'm checking against instead of just looking around. I compare what's actually happening at the station to what the documents say, and I check that calibration stickers and gauge records are current. On my last internal audit, I found an operator using a work instruction revision that was 2 revisions old. I wrote it up as a minor nonconformance, and the corrective action was a control plan review step whenever engineering issues a new revision, not just blaming the operator.
14. What measurement system analysis do you run before you trust a gauge?
Why they ask: Bad data from an unreliable gauge can hide a real process problem or create a false one.
How to answer: Mention Gage R&R, the acceptance thresholds you use, and what you do when a study fails.
Sample answer: I run a Gage R&R study with 3 operators, 10 parts, and 3 trials each before I trust a new or disputed gauge, and I look at percent of total variation from repeatability and reproducibility, targeting under 10% as acceptable and flagging anything over 30%. Last year a height gauge failed with 42% gauge R&R, mostly reproducibility between operators. A fixture that removed the operator's judgment call on where to seat the part brought the study down to 8% on the retest, without replacing the gauge itself.
Behavioral questions
15. Tell us about a time you found a root cause that wasn't the obvious one.
Why they ask: They want proof you dig past the first plausible explanation.
How to answer: Show the obvious suspect, why you kept looking, and the actual cause.
Sample answer: A customer reported intermittent electrical shorts on a connector assembly, and everyone assumed it was a supplier's wire insulation. I pulled 15 returned units and found the shorts clustered near one crimp station, not randomly across suppliers. I checked the crimp height data and found the crimper's die had worn beyond spec, something the daily setup check wasn't catching because it only verified pull force, not crimp height. Adding a height check to the setup sheet caught the same wear pattern on a second crimper before it shipped a single bad part.
16. Describe a time you had to place a quality hold or stop production.
Why they ask: This tests whether you'll act on a real risk even when it's unpopular.
How to answer: Describe the risk, the decision, and how you communicated it.
Sample answer: During a shift change, I noticed a batch of torque readings on a fastener station trending toward the low spec limit, though none had failed yet. I placed a hold on about 1,200 parts made since the last verified check and asked the supervisor to stop the line for a torque wrench calibration check. The wrench was 8% out of calibration. The plant manager wasn't happy about a 40-minute stoppage, but re-checking the held parts found 60 below minimum torque, which we caught before they left the building instead of after a customer complaint.
17. Tell us about a time you disagreed with a manager or engineer about a quality decision.
Why they ask: Quality engineers need to hold a position under pressure without becoming impossible to work with.
How to answer: State the disagreement plainly, how you supported your position with data, and the outcome.
Sample answer: An engineering manager wanted to ship a lot of parts with a cosmetic scratch outside the print's surface finish callout, arguing the customer wouldn't notice. I disagreed, since the print didn't have a cosmetic waiver clause and the part was a visible exterior surface. I brought him the actual print requirement and a photo comparison against the approved sample, then offered a deviation request to the customer as the correct path instead of shipping without one. He agreed to submit the deviation, and the customer approved it within a day with a note in our file, which kept us clean on the next audit.
Situational questions
18. A customer returns a batch with a defect your inspection plan should have caught. What do you do?
Why they ask: This tests your response under customer pressure, not just your technical skill.
How to answer: Cover immediate containment, root cause on both the defect and the inspection gap, and customer communication.
Sample answer: First I'd contain: sort remaining inventory at our plant and in transit, and ask the customer for return samples so I can see the actual failure mode instead of a description. Then I'd run root cause on 2 tracks: why the part failed, and why our inspection plan didn't catch it, since those can have different answers. If the inspection sampling plan missed it because of a low sample rate, I'd fix the plan and say so plainly in the 8D, rather than only addressing the part defect and leaving the same detection gap open for the next failure mode.
19. You're 2 days from a PPAP submission deadline and a capability study fails. What do you do?
Why they ask: They want to see whether you'd hide a bad result or handle it correctly.
How to answer: Say you'd investigate the failure, not force a passing number, and communicate the delay early.
Sample answer: I'd tell the program manager and my quality manager the same day I see the failed Ppk, since a surprise 2 days out is worse than an early flag. Then I'd dig into why it failed: a tooling issue, a process parameter that needs adjustment, or a spec that's tighter than the process can hold. On a similar situation last year, the cause was a fixture locating surface with too much wear, and correcting it took 4 extra days. I'd rather submit a PPAP 4 days late with real data than submit on time with a capability study I don't trust.
20. An IATF 16949 auditor flags a nonconformance in your CAPA system. How do you respond?
Why they ask: How you handle audit findings shows whether you treat the QMS as real or as paperwork.
How to answer: Describe reviewing the finding objectively, building a real correction and root cause response, and the timeline auditors expect.
Sample answer: I'd read the finding carefully and confirm I understand exactly what evidence the auditor is pointing to, since a vague response to an unclear finding gets rejected. Most certification bodies expect a containment response within a few days and a full root cause and corrective action plan within 30 to 60 days, so I'd start the root cause work immediately rather than waiting. On a finding I closed last year, about CAPA effectiveness checks not being documented, the fix was adding a mandatory 90-day effectiveness review field to our CAPA template, which the auditor verified and closed on the next surveillance audit.
Questions to ask the interviewer
- What quality management software and SPC tools does the team use day to day?
- Is this plant IATF 16949 certified, ISO 9001 only, or working toward certification?
- How many active CAPAs or open nonconformances does the quality team typically manage at once?
- What does a new part launch look like here, and how early does quality get involved in APQP?
- Who does this role report to, and how much authority does a quality engineer have to stop the line?
- What's the biggest recurring quality problem the team hasn't solved yet?
How to prepare
- Review your SPC math. Be ready to calculate Cp, Cpk, and Ppk from sample data without a calculator app doing the thinking for you.
- Bring a real root cause story. Have 2 or 3 investigations ready with actual numbers: defect rates, RPN scores, or capability indices.
- Refresh PPAP and APQP details. Know the 5 PPAP levels and the 5 APQP phases well enough to explain them without notes.
- Check the employer's certifications. Look up whether the plant is IATF 16949, ISO 9001, or AS9100 certified, and tailor your audit and CAPA answers to that standard.
- Practice explaining FMEA scoring. Be ready to walk through severity, occurrence, and detection on a specific failure mode, not just the definition.
- Prepare for a data exercise. Some interviews include a live dataset or a nonconformance report to analyze on the spot.
If you're also interviewing for adjacent roles, these quality control lab technician interview questions cover similar inspection and documentation skills, and operational excellence manager interview questions go deeper on process improvement leadership. For supply-chain-adjacent roles that intersect with PPAP and supplier quality, see procurement specialist interview questions. Engineers moving from a different discipline may also find chemical engineer interview questions and RF design engineer interview questions useful for comparing how technical interviews are structured in other engineering tracks.