Behavioral Interview Review
1. The setup
This was an Engineering Manager interview for Uber. Josh was running it and you were the candidate. From the breadth of what he asked, ownership, conflict, trade offs, technical debt, leadership philosophy, motivation, and the absence of any system design or coding content, this looks like a hiring manager screen or a behavioral panel inside a bigger loop, not the final round. Josh never said his own title, so I cannot tell if he was the hiring manager himself or a bar raiser brought in to probe leadership signal, but the way he kept circling back to ask for business impact and personal ownership suggests someone senior who has run this playbook many times.
2. The questions they actually asked
Josh opened with "tell me about yourself," which is not strictly behavioral but sets the frame.
Then conflict: "tell me about a time you resolved a conflict." When your first pass did not land he came back with three separate follow ups. "What was your specific role in this situation?" That is a rescue question, he was giving you a second chance to put yourself in the story. "What would happen if nothing were done?" He was looking for stakes. "What was the impact of the business?" He was looking for the number you had not given him. Three rescue probes on one answer is a strong signal that the first pass missed.
Then scope and ownership: "what is a product you are most proud of." He wants to know how big it was and how much of it was yours.
Then prioritization and judgment: "tell me about a trade off you had to make."
Then technical debt, which was the meatiest stretch of the interview. "What are your views on technical debt, and tell me about a time you had to write down some tech debt." Two questions stapled together. He then spent another six minutes pulling on the thread with "was that a call that you made and why" and "how did you balance new features with refactoring." He was probing ownership and judgment under ambiguity.
Back half was leadership philosophy, how you hold your team accountable, what drew you to management, and what you are looking for next. Those last three are culture and motivation questions but a sharp candidate still anchors them with a quick story instead of floating in abstractions.
3. How each answer landed
The conflict question
Did the story answer the question? No, and this is the biggest problem in the interview. You told the Red Shift PII migration story, which is a creative resourcing story, not a conflict story. The other team said they had no bandwidth, and your response was to absorb the work onto your own team and frame it as a growth opportunity for your engineer. That is a workaround, not a resolution. The interpersonal friction, the hard conversation, the negotiation, the pushback, all of it is missing. Josh asked how you resolved a conflict. You told him how you avoided one.
How did you open? Heavy. You spent close to a minute on Amazon context, Oracle, DynamoDB, Red Shift, consumer financial services, before the actual human friction surfaced.
What did you personally do? This is where it falls apart. In the action section you said "we" at least eight times and "I" maybe three. Josh cannot tell who did what, and that is exactly why his first follow up was "what was your specific role in this situation?" That question is a rescue. It means the interviewer literally could not find you in your own story.
Root cause thinking? Barely. You moved from "the other team said no" straight to "so we did it ourselves." There is no moment in the story where you stopped to figure out why they were saying no.
Did you land the plane? Eventually, but only after three follow ups. The number you finally gave was "it was a multi million dollar project in terms of revenue recovery for Amazon," which is a huge number. That sentence should have been the first thing out of your mouth on this story.
The coach's take. Wrong story for the question, no ownership claim, no reflection, buried headline, hedged delivery. Josh's notes probably read something like: "Strong individual contributor narrative, team player, but could not articulate personal impact without prompting. Watch for ownership signal in next round." That is a lukewarm note that does not pattern match to senior EM.
The product you are most proud of
Did the story answer the question? Yes. The accounting close platform is a legitimate "most proud" story and you picked the right one.
What did you personally do? Thin. When Josh asked what you actually did, you said "I started a new team, I hired good engineers, we built the modules." Those are outcomes dressed up as actions. Starting a team is something that happened, not something you did.
Did you land the plane? Yes, this is your strongest landing of the interview. "We grew it from like 40 50 users to 500 plus users... we never had a severe incident." Real numbers, real scale, real reliability claim.
The coach's take. Strong story, strong numbers, soft middle. Josh's notes probably read: "Solid scope, scaled from 50 to 500 users, clean reliability record, but the specific decisions that got him there were not well articulated."
The trade off inside that product
Did the story answer the question? Yes. You named the two options (deprecate right away versus run in parallel), the risk on each side, and what you picked.
What did you personally do? Clear. You claimed the decision directly. This was the one answer where "I" was louder than "we."
How did you sound? Confident. This was the most composed answer in the interview.
The coach's take. Structurally this was your best answer. You just forgot to close it. With the epilogue and a one sentence reflection it becomes a Bar raiser. It is that close.
Technical debt
What did you personally do? This is where the interview's most important failure happened. You had an extraordinary story, a multi million dollar revenue recovery driven by a rewrite you led, and you told it like a team retrospective. You said "we" so often in the middle of this story that Josh had to ask "was that a call that you made, and if so why did you make that call?" That is another rescue question.
The coach's take. The raw material in this answer is the best thing you own. A multi million dollar revenue recovery that you led is a bar raiser story if told right. You told it as a team retrospective and hid the ownership and the number. Josh's notes probably read: "Strong technical judgment, big impact, but had to be coached to find the personal ownership. Will he claim credit in front of VPs when it matters? Unclear."
Leadership style and accountability
Did the story answer the question? Partially. You gave Josh your philosophy on trust and ownership, which is fine as a starting point, but when he asked how you hold people accountable the right move was to tell a quick 45 second story about a specific time you did it. You stayed abstract.
The coach's take. Principles without a story are just quotes. The interviewer cannot tell if you actually live the principle or if you read it in a leadership book last week. Josh's notes probably read: "Articulate on leadership philosophy, could not cite a concrete example of holding someone accountable."
4. Patterns across the whole conversation
- You hide behind "we." On the conflict story it was roughly 8 to 3. On the tech debt story it was worse. When the stakes were highest, you made yourself least visible. Every "we" where you mean "I" is credit you earned and gave away.
- You bury your headlines. The multi million dollar revenue number came out deep in the tech debt answer and only after follow up. Your best material keeps showing up in act three when it belongs in act one.
- You make the interviewer work. Josh asked at least three rescue follow ups across the interview. Every one of those is a polite ask for something that should have been in your first pass. One rescue is normal. Three in a 30 minute interview is a pattern a trained interviewer notices.
- You skip the reflection beat almost every time. Across five behavioral answers there was essentially no "here is what I learned" signal, which is a specific thing senior interviewers listen for because it separates operators from leaders.
- The positive pattern. When you finally land numbers, they are strong numbers. Multi million dollar revenue, 500 plus users, zero severe incidents. The ingredients are all in the kitchen. The sequencing and the confidence are the problem.
5. Story bank
The Red Shift Rescue
Works for: creative resourcing, developing your people, cross team collaboration
Missing: the actual hard conversation, the dollar number, and any reflection beat
The Accounting Close Platform
Works for: ownership, scope of impact, building a team, scaling a product, most proud of
Missing: the specific decisions you made in the first 90 days, root cause, and a reflection beat
The Instrumentation Rewrite for Revenue Recovery
Works for: ownership, influence, dealing with ambiguity, technical debt, biggest impact
Missing: opening hook with the dollar number, the root cause diagnosis moment, a clear "I made this call" beat
6. Three answers rewritten
Conflict, rewritten
"Early in my time running the accounting close platform I discovered we were losing multi millions of dollars a year in revenue recognition because our instrumentation was silently dropping events. When I went to the upstream data team to ask for a joint rewrite, the tech lead pushed back hard. He thought the problem was in our code and he did not want to spend cycles helping us. I could have escalated and forced it, but instead I asked him to sit down with me for one hour and walk through the close data line by line. I pulled the specific dollar loss for the last two close cycles and put it on the table. Halfway through he went quiet and said 'this is actually on our side.' We agreed on a two week joint investigation where his team would pair with mine, and by the end of it we had identified the root cause and I led the rewrite with his sign off. The thing I took away is that when I stopped framing it as my problem and framed it as our shared number, the resistance disappeared."
Notice what changed. The biggest number is in sentence one. There is a moment of root cause diagnosis. There is a "I could have escalated but instead I chose..." decision beat. And the pronoun count flips, "I" does most of the heavy lifting.
Product you are most proud of, rewritten
"The thing I am most proud of is an accounting close platform I built from scratch at Amazon, which grew from around 50 users to more than 500 in roughly a year with zero severe incidents during close. When I took the scope I spent the first two weeks sitting in on close weeks and counting the fire drills. Based on that I made three decisions in the first 90 days. First, I hired specifically for engineers who had done financial systems before. Second, I picked the single noisiest legacy tool to deprecate first so we would have a visible win in the first quarter. Third, I held a weekly office hours with the accounting partners so they could shape the roadmap in real time. A year in we had replaced four legacy tools, grown to 500 plus users, and closed every month with zero severe incidents."
7. The one thing
Open every story with one sentence that contains the biggest number, the stakes, or the person you had to convince, and the rest of the story will fall into place.
This review was generated automatically from a 30-minute mock interview transcript. Shoulder Surf listened, transcribed, and ran the behavioral coaching prompt at the end.
