Microsoft Interview Guide 2026: Model-Coach-Care, As Appropriate, Growth Mindset

Company Guide · Updated April 2026 · Reviewed by a former Microsoft principal engineer (Azure, 7 years)

Microsoft's behavioral rubric is one of the few fully public ones in big tech — printed on the careers site, drilled into every interviewer in training, cited in every performance review. And almost nobody preps for it explicitly. Candidates walk in ready for FAANG-style questions, get asked a Model-Coach-Care (MCC) behavioral, and give an answer that would have scored at Google or Amazon but reads flat against Microsoft's rubric.

This guide reads the MCC rubric end-to-end with scored examples for each pillar, unpacks "as appropriate" (the scope signal that separates senior from principal), and shows how growth mindset stories actually land with interviewers. If you're comparing big-tech loops in parallel, pair this with our Amazon Leadership Principles guide, Google Googleyness guide, and Meta cultural-bets guide. The behavioral interview guide covers STAR mechanics.

The MCC rubric in plain English

Model-Coach-Care is Microsoft's three-pillar rubric for every hiring and performance-review decision. The language is public and consistent:

• Model — Set direction and boundaries, bring clarity to others, generate energy.
• Coach — Grow the capabilities of others. Develop people, not just output.
• Care — Contribute to the success of others beyond your role. Show up for the team, the org, and the customer.

Every behavioral question you're asked in a Microsoft loop maps to one of these three — and the interviewer's note will cite the pillar explicitly. "Candidate demonstrated Model through…" or "Care example was shallow — asked for follow-up." If your answer doesn't connect clearly to one of the three, the interviewer has to write "non-example" in the rubric cell, which is a down-vote.

The implication: while you prepare stories, tag each one with its primary pillar. If you don't know which pillar a story serves, the interviewer won't either.

Setting direction (Model)

Model is the pillar most candidates under-prepare for because it sounds like generic leadership. It is not. Microsoft's rubric scores Model on three specific behaviors:

• Clarity generation. Took a fuzzy situation and made it legible for others (a design doc, a scorecard, a decision memo).
• Boundary setting. Said no to a scope expansion, a premature launch, or a competing priority — and preserved the relationship.
• Energy creation. Your team worked harder or better because you were in the room, not in spite of it.

A scored Model answer at the senior-engineer level:

Prompt: "Tell me about a time you set direction for a team."

Answer: "Our service had three concurrent migrations in flight — observability, auth, and database — and the team was exhausted. I wrote a one-page memo ranking the migrations by customer risk, got sign-off from my manager and the two partner team leads in 48 hours, and paused the two lower-priority migrations for a quarter. The observability migration shipped on time, the paused migrations re-started clean in the next quarter instead of limping on, and two engineers who had asked to move teams stayed. Productivity as measured by our flow metric (code-review turnaround) recovered to pre-migration levels within three weeks."

Notice what the rubric cell gets to cite: a written artifact (the memo), a boundary held (two migrations paused), a named metric that moved (flow), and a retention outcome. Every Model story should give the interviewer three or four citable moments.

Growing others (Coach)

Coach is where candidates over-invest in their reporting-line stories. Microsoft's rubric rewards both: coaching you did as a manager and coaching you did as a peer. For non-manager candidates, peer coaching is actually the stronger signal because it's harder to evidence.

The Coach pillar scores:

• Named development goals for the person you coached.
• A concrete mechanism you set up — 1:1 cadence, code-review pairing, a growth plan, a rotation.
• A measurable outcome for the person (promotion, new scope, unblocked project) and for the org (retention, delivery).

A scored Coach answer:

Prompt: "Tell me about someone you developed."

Answer: "A mid-level engineer on my team wanted to move into system design but had never written a design doc that aligned three teams. I paired with them on a redesign of our rate-limiter — they owned the doc, I reviewed each draft for 30 minutes weekly. Over six weeks the doc went from 'my current understanding' to a signed-off proposal. They led the implementation, presented the results at the team all-hands, and were promoted at the next calibration — two cycles earlier than the team average. They later led two design reviews for newer engineers using the same pairing cadence."

The rubric cell can cite: the explicit goal (system design), the mechanism (weekly 30-min reviews), the outcome for the person (early promotion), and the downstream impact (they now coach others — Care meets Coach).

Contributing beyond your own success (Care)

Care is the pillar most often thinned to "I'm a team player." That phrase scores zero. The Care rubric is specifically about times your own work was deprioritised in favour of helping someone else succeed.

What Care asks you to evidence:

• You chose to help someone else ship, even when it cost your own delivery.
• You invested in the org's health (oncall quality, onboarding docs, hiring rubric) without being asked.
• You made a customer successful whose contract didn't cover your work.

A scored Care answer names the specific trade. "I spent three weeks improving our onboarding because two engineers had quit in their first 90 days — my Q3 feature slipped by a sprint, but we haven't lost a new hire in six months since." The rubric cell gets to cite the trade, the metric, and the durable improvement.

"As appropriate" — the scope signal

Every MCC rubric cell ends with the phrase "as appropriate for the level." It reads like filler; it's the scope signal the interviewer uses to calibrate. The same behaviour demonstrated at junior scope is a "meets" at IC2, a "below" at Senior, and a "below expectation" at Principal.

Calibrate your stories. If you're interviewing for Senior (IC4), the interviewer expects:

• Model stories that align two or more teams, not just your own.
• Coach stories that span at least two quarters of development.
• Care stories where the trade you made was visible to an org — a skip-level or partner team noticed.

If your strongest story is a single-team, single-quarter Coach, find a second story at broader scope before the loop. Microsoft's hiring committee will flag a debrief where every scored example is individual-scope as "did not meet As-Appropriate at the target level."

Growth mindset stories that land

Every Microsoft loop includes at least one Growth Mindset question — a cousin of Are-Right-A-Lot from Amazon's rubric. The shape: "Tell me about a time you were wrong."

What scores well:

• The mistake is named concretely (a decision, a call, a prediction — not "I didn't manage my time").
• You name what changed in your process as a result. Not a resolution ("I'll be more careful"), a mechanism ("I now run a pre-mortem before any migration").
• The mechanism outlived the story — you can cite a later situation where the new process paid off.

What scores poorly: humblebrags ("I was too ambitious"), abstract lessons ("communication is important"), or a mistake that was someone else's in disguise.

Prepare two Growth Mindset stories. The first should be the cleanest mistake you ever made. The second should be a mistake that is still slightly uncomfortable to tell. Interviewers listen for the discomfort — it's the tell that the story is genuine.

Connect framework for performance reviews (and why it matters for interviews)

Microsoft runs performance reviews with a framework called Connect. Every review names three Connects:

1. Individual accomplishments ("What did you deliver?")
2. Contributions to others' success ("How did you help teammates?")
3. Contributions that built on others' work ("What did you learn from and build on?")

The three Connects map one-to-one to Model, Coach, Care. The implication: your interviewer is in the same framework five days a year themselves. A story that maps cleanly to the three Connects reads as native. A story that doesn't — for instance, an entirely individual accomplishment with no mention of what you built on — reads as misaligned.

If your prep spreadsheet tags each story with its primary and secondary pillars, and at least one story names what you built on from another team, you'll sound like someone who would pass their first Connect review. That's the shape Microsoft's interviewers are hiring for.

Frequently Asked Questions

What is Microsoft's MCC rubric?

Model-Coach-Care is Microsoft's three-pillar behavioral rubric used in both hiring and performance reviews. Model is setting direction and generating clarity. Coach is growing others' capabilities. Care is contributing to the success of others beyond your role. Every behavioral question in a Microsoft loop maps to one of the three.

How many interviews are in a Microsoft loop?

Four to five onsite rounds: two technical (coding or system design, depending on level), one behavioral, one "As-Appropriate" scope round, and one skip-level or cross-team round for senior candidates. A recruiter screen and a technical phone screen precede the onsite.

How is Microsoft's behavioral interview different from Amazon's?

Amazon maps every behavioral answer to one of 16 Leadership Principles; Microsoft maps to three MCC pillars. Amazon's Bar Raiser drills one story for 45 minutes; Microsoft's interviewers run two to three shallower stories and score each against the relevant pillar. Microsoft weights "as-appropriate" scope calibration more explicitly than Amazon does.

What is growth mindset at Microsoft?

Growth mindset is Microsoft's behavioural signal for intellectual humility and learning from failure, a Satya Nadella-era cultural anchor. Interview scoring rewards specific mistakes with a named process change that outlived the story, not abstract lessons or rebranded humblebrags.

Do I need one story per MCC pillar?

At minimum one per pillar, ideally two. Microsoft loops often ask two behavioral questions mapped to the same pillar back-to-back (two Coach, for example) — one story per pillar leaves you reusing, which the rubric explicitly penalises as lack of depth.

What level is Senior at Microsoft?

Senior is typically level 63 (individual contributor) or 64, depending on the org. It is the terminal level for strong ICs who don't pursue Principal. Expected scope at Senior includes leading design reviews for peers, owning a multi-quarter initiative, and setting direction for at least one cross-team decision per year.

Keep reading

• The Behavioral Interview Guide: STAR, Stories, and How to Actually Win — the pillar guide
• Amazon Leadership Principles Interview Guide (2026)
• Google Behavioral Interview Guide: Googleyness Explained (2026)
• Meta Interview Process 2026: Loops, Rubric, E4–E6 Prep Guide

Ready to drill MCC-tagged stories against the As-Appropriate rubric? Start a free trial — Microsoft-preset questions with pillar scoring and level calibration included.