How to Hire Engineering Managers in 2026

Engineering manager hiring is the most consequential and most poorly-run process in most companies. A bad IC hire costs you a quarter of low-velocity work. A bad EM hire costs you a year of attrition, missed deadlines, and the energy you'll spend covering for someone who can't do the job. Yet the typical EM loop borrows directly from IC loops — a coding screen, a system design round, a take-home, three behavioral interviews — and reliably surfaces the wrong signal.

This playbook is for heads of engineering, founding engineers, and talent partners who hire managers and want a process that actually works in 2026. It draws on patterns we've seen across the 118 companies in our directory and what high-quality hiring teams are doing differently. It's opinionated — if you want a generic checklist, this isn't it.

Why EM Hiring Usually Fails

The EM hiring loop most companies run was designed for IC hiring with a few "leadership" questions bolted on. That format optimizes for the wrong things. Strong managers are not strong because they can solve a coding problem in 45 minutes — they're strong because they can read a team, set direction, navigate conflict, and make calibrated decisions with incomplete information. None of those show up in a coding screen.

The default loop also confuses interview-performance with job-performance. A polished candidate who has interviewed at twenty companies and rehearsed every behavioral question can sail through the loop. They may also be a terrible day-to-day manager. Meanwhile, a brilliant operator who has been heads-down running a team for three years and forgot how to "tell me about a time" can crash the interview and walk away from a role they would have crushed.

The third failure pattern is the "previous EM archetype" trap. The team had a great EM who was deeply technical. That EM left. Now hiring is screening for "deeply technical EM" — but the team's actual problem in the next 18 months is that it's growing from 8 to 20 people and needs a strong people-development manager, not another technical lead. The previous EM solved last quarter's problem. Hiring needs to solve next year's problem.

Step 1: Define What This Specific Team Needs Now

Before you write a job description, spend 30–60 minutes writing what the team will actually need over the next 18 months. Be specific. Vague answers like "strong engineering leader" guarantee you'll hire whoever charms the loop the most. Strong answers give you something to test for.

The questions to answer:

• What's the biggest open problem on this team right now? Is it shipping velocity? Hiring? Performance management of underperformers? Cross-functional alignment? Tech debt? Each of these calls for a different EM.
• What size will the team be in 12 months? Managing 5 people is different from managing 12. If the team is going to double, you need someone who has done that growth journey before.
• How technical does the EM need to be? Hands-on code-reviewing technical, or strategic-decisions technical? Get specific.
• What's the team's relationship to product? An EM in a product-led org runs differently from an EM in an engineering-led org. See the difference between engineering-driven cultures and product-impact cultures.
• What kind of stakeholder management does the EM need to do? Manage up to a VP? Coordinate with two sister teams? Talk to customers? Each is a different skill.

Once you've answered these, you have a job description that's actually testable. Now the loop can be designed against the answers.

Step 2: The Four-Stage Loop

The loop has exactly four stages. Each tests one thing and tests it well. Total candidate time is roughly 4 hours plus a 90-minute take-home-ish exercise. Total calendar time is 4–6 weeks.

Stage 1: Recruiter Screen (30 min)

Standard logistics: timeline, comp, geography, motivation for change. Two specific things to add for EM hiring: ask why they're leaving their current role (the answer tells you whether they're being pulled or pushed), and ask how many direct reports they currently have. EMs sometimes round up generously.

Stage 2: Hiring-Manager Judgment Interview (75 min)

This is the most important interview in the loop. You (or whoever the EM will report to) walk through three concrete situations from the candidate's last two years and ask, for each: what was happening, what did you do, what would you do differently. The goal isn't to test their narrative skills — it's to assess judgment, self-awareness, and decision quality.

Strong candidates will tell you specifically what they did, why, what trade-offs they weighed, what they got wrong in retrospect. Weak candidates will tell you about the situation in detail and stay vague about their own role and reasoning. The signal is in how much "I decided" and "I should have" shows up in their answer.

Three situations to ask about, calibrated to your team's needs:

• A time they had to deliver bad news to a strong performer (skill: performance management, courage)
• A time they had to make a call with incomplete information (skill: judgment under uncertainty)
• A time they had to push back on a senior stakeholder (skill: managing up, technical credibility)

Stage 3: Peer Panel With Two ICs (60 min total, 30 min each)

The candidate meets two ICs from the team they would manage. Not to be evaluated, but to evaluate the candidate. ICs ask their own questions. They report back to you afterward with a thumbs up, thumbs sideways, or thumbs down, and a paragraph of reasoning.

This stage does two jobs. It surfaces whether the candidate listens, respects engineers, and can switch from talking strategy to talking craft. And it gives your team agency in their next manager — which dramatically improves the new manager's onboarding success. Engineers who interviewed their new EM and gave them a thumbs up have already made an emotional commitment to making the relationship work.

Stage 4: 90-Day Plan Exercise (90 min review)

This replaces the system design round and the take-home. Send the candidate a redacted summary of the team's current state — size, roadmap, two or three open problems, recent wins, recent losses. Ask them to come back in a week with a 90-day plan: what they'd focus on, what they'd defer, what they'd want to understand better, and what their first one-on-one with each direct report would look like.

You're not looking for the right answer — there isn't one. You're looking for how they think about prioritization, what they notice, what questions they ask, and how they balance hands-on engagement with structural decisions. A weak candidate will give you a generic "listen for 30 days then decide" answer. A strong candidate will tell you specifically which decisions they'd want to make on day 1 versus day 30 versus day 90, and why.

Step 3: Reference Checks (The Highest-Signal Step)

Most companies do reference checks as a formality. Don't. For EM hires, references are the highest-signal step in the whole process — higher than any interview. Block 90 minutes for references after the loop and before the offer.

Ask the candidate for two engineers and one peer manager from their last role. Call them directly. Email is fine to schedule but the conversation has to be live, because the most useful signal is in tone, pauses, and what they don't say.

Three questions that work:

1. "What would you specifically want to know if you were considering having this person manage you?" — surfaces concerns gracefully.
2. "On a scale of 1–10, how excited would you be if they became your manager again?" — anything under 8 deserves follow-up. Even an "8" from someone who pauses first is meaningful.
3. "Where do they need development?" — honest references give you a specific weakness. Polite references give you a strength reframed as a weakness ("they care too much"). The difference is the signal.

What to Stop Doing

A few things that show up in default EM loops that you should drop:

• System design rounds. Useful for IC hires. Useless for managers above ~5 reports. If technical credibility matters, test it in the judgment interview by asking about a specific technical decision they made and how they thought about it.
• Roleplay exercises ("pretend I'm your underperforming direct report"). Reward acting ability, not management ability. Replace with concrete past situations.
• "Tell me about a time you failed." Everyone has rehearsed an answer. Replace with "What's the call you've made in the last six months that you'd most want to take back?" — same question, much harder to fake.
• Eight-person interview loops. They optimize for false negatives (someone gets a low score somewhere) and exhaust strong candidates who have multiple offers. Four interviews is enough.
• Take-home exercises lasting more than 90 minutes. Senior candidates won't do them. The ones who will are usually less competitive.

Internal vs External: When to Promote and When to Hire

Promoting from within and hiring externally both have failure modes. The right answer depends on what the team needs.

Promote internally when: the team has a strong senior IC who has been informally leading for 6+ months, the team needs continuity, and the company has investment in management training. The risk: an IC promoted to EM without training will struggle for 6–12 months while the team's velocity drops.

Hire externally when: the team needs a step-change (e.g., scaling from 5 to 20), no internal candidate has the experience pattern, or political dynamics inside the team make any internal promotion contentious. The risk: external EMs take 3–6 months to learn the politics and culture, and the team's trust has to be earned.

Combine both when: the team is large enough to support a manager and a deputy. Promote the high-potential internal IC to a tech-lead or staff-engineer role with people responsibilities. Hire the senior EM above them. The internal person gets a credible growth path. The team gets continuity plus the experienced operator. This is the move most under-used by mid-stage companies.

How to Sell the Role to Strong Candidates

Strong EM candidates have options. They are interviewing you the whole time. Four things they're actually looking for:

1. The specific problem the team is trying to solve. Generic mission language ("change the world", "the future of work") doesn't move EMs. Specific problems do. "We have to triple infrastructure capacity in 9 months while staying under a fixed compute budget" is a problem that excites operators.

2. The engineers they'd manage. Make sure they meet 2–3 of them in unstructured conversations, not just the formal interview panel. EMs care a lot about who they'd be working with day-to-day.

3. The leverage they'd actually have. Budget, hiring authority, roadmap influence, ability to say no to bad ideas. EMs have been burned by hidden constraints in previous jobs. Be specific about what they own and what they don't.

4. The engineering culture and values. Strong EMs read engineering culture like a contract. Companies that publish their culture clearly — like Stripe, Linear, and Vercel — close EM candidates faster than companies whose careers page reads like marketing copy.

What the First 30 Days Should Look Like

Hiring is half the work. The first 30 days are the other half. Before the candidate's start date, decide:

• Who will they have a weekly 1:1 with? Their manager, yes. Also: one senior IC on the team they manage, and one peer EM.
• What's the one decision you'd like them to make by day 30? Naming it surfaces whether they have the latitude they expected.
• Who is responsible for telling them about the political landscape? Don't make them learn this from incidents. Assign someone to brief them in week one.
• What can they read? Internal docs, last quarter's planning, recent incidents, last performance review cycle's outcomes. Hand it to them.

The new EM's first month is largely about earning trust. The structural decisions you make about how to set them up determine how fast that trust accrues.