Intervu is in beta — feedback welcome at support@intervu.io

GitLab DevOps / SRE Interview Questions

30 real practice questions for the mid-level DevOps / SRE role at GitLab (Developer Tools), spanning behavioral, technical, system design, leadership, and problem solving. Build and maintain infrastructure, CI/CD pipelines, and ensure system reliability. The first 3 questions below include what GitLab interviewers actually listen for, plus likely follow-ups.

Practice these questions with AI feedback
Questions
30
Categories
Behavioral (6), Technical (6), System Design (6), Leadership (6), Problem Solving (6)
Difficulty mix
10 easy · 10 medium · 10 hard
Avg. answer time
~4 min

Behavioral Questions (6)

  1. 1.Tell me about a time you had to document a complex process or incident resolution to make it accessible to your team. What did you document and how did you structure it?

    easy~3 min

    What interviewers look for

    • Created comprehensive, searchable documentation that became a reference for future incidents or onboarding
    • Made the documentation public or easily accessible to all team members, not just immediate stakeholders
    • Used a structured format with clear sections, troubleshooting steps, and ownership details
    • Updated the documentation iteratively based on feedback or new learnings

    Likely follow-ups

    • How did you decide what level of detail to include in the documentation?
    • What feedback did you get from team members who used this documentation later?
    • How do you keep documentation like this up to date as systems evolve?

    Company context

    GitLab operates under the 'Handbook First' principle where everything is documented publicly by default. For DevOps engineers managing complex infrastructure, this means incident responses, runbooks, and system knowledge must be thoroughly documented so any team member can contribute or respond to issues. This reflects GitLab's transparency value and all-remote culture where written communication is essential.

  2. 2.Walk me through a time you had to create documentation or a runbook for a critical system failure so others could handle similar issues. What was your approach to making it comprehensive yet usable?

    easy~3 min

    What interviewers look for

    • Created actionable documentation that enabled others to respond to similar incidents without escalation
    • Included both immediate response steps and background context for understanding the issue
    • Made the documentation easily discoverable and accessible during high-stress situations
    • Tested the documentation with team members or used it themselves in subsequent incidents

    Likely follow-ups

    • How did you balance providing enough detail without overwhelming someone in a crisis?
    • What format or tools did you use to make the runbook accessible during incidents?
    • How do you ensure runbooks stay current as systems change?

    Company context

    GitLab's 'Handbook First' principle requires that critical operational knowledge be documented publicly and accessibly. For DevOps engineers managing GitLab's infrastructure, this means incident response procedures must be clear enough for any team member to follow. This supports GitLab's all-remote culture where knowledge can't be tribal and reflects the transparency and efficiency values.

  3. 3.Describe a time when you had to choose between implementing a new, exciting technology and sticking with a proven, boring solution for a DevOps challenge. Walk me through your decision process.

    medium~4 min

    What interviewers look for

    • Explicitly chose the boring solution over the shiny new technology based on practical considerations
    • Evaluated factors like team expertise, maintenance burden, and risk of the new technology
    • Considered the opportunity cost and timeline implications of adopting new tech
    • Communicated the decision clearly to stakeholders who may have preferred the exciting option

    Likely follow-ups

    • What was the new technology that you decided against and why was it tempting?
    • How did your team react to choosing the boring solution?
    • Looking back, do you think you made the right call? What would you do differently?

    Company context

    GitLab's 'Boring Solutions' principle emphasizes choosing proven, simple approaches over exciting new technologies. For DevOps engineers managing GitLab's infrastructure at scale, reliability and maintainability trump innovation. This principle reflects GitLab's efficiency value and focus on results over activity - the best solution is often the one that just works.

  4. 4.Tell me about a time you identified and fixed a problem outside your direct area of responsibility. What was the issue and how did you approach getting involved?

    medium~4 min

  5. 5.Describe a situation where someone from outside your team made changes or contributions to your infrastructure or monitoring setup. How did you handle their involvement?

    hard~5 min

  6. 6.Tell me about a time you had to resist pressure to implement a cutting-edge technology or tool that your team was excited about. What was the technology and how did you navigate the situation?

    hard~5 min

Technical Questions (6)

  1. 7.A team deployed a change that increased Redis memory usage by 40% on GitLab.com. They want to roll forward with additional optimizations rather than roll back. How do you handle this situation?

    easy~3 min

  2. 8.GitLab's Kubernetes clusters need to be upgraded across multiple GCP regions, affecting CI/CD runners and the main application. How would you plan and execute this upgrade while minimizing user impact?

    easy~3 min

  3. 9.GitLab.com processes over 300,000 CI/CD jobs daily across thousands of runners. You notice that job queue times have increased from 30 seconds to 5 minutes during peak hours. How would you investigate and resolve this?

    medium~4 min

  4. 10.You need to set up monitoring for a new Go microservice that handles GitLab's dependency scanning results. What metrics and alerts would you implement, and how would you structure them?

    medium~4 min

  5. 11.You're oncall and GitLab.com's PostgreSQL primary is showing 95% CPU utilization with slow query warnings. The incident is affecting merge request creation. Walk me through your immediate response.

    hard~5 min

  6. 12.A product team wants to add real-time collaboration features to GitLab's Web IDE. This would require WebSocket connections and could affect our current Rails architecture. How would you approach this technical decision?

    hard~5 min

System Design Questions (6)

  1. 13.Design a pipeline executor service that can handle 500,000 CI/CD jobs per day across GitLab's runner fleet. How would you ensure jobs are distributed efficiently and handle runner failures?

    easy~3 min

  2. 14.Design a merge request notification system for GitLab that can handle a repository with 10,000 watchers where a single MR update could trigger millions of notifications. How do you prevent overwhelming users and infrastructure?

    easy~3 min

  3. 15.GitLab Duo's AI features need to serve code suggestions to millions of developers with sub-200ms latency. Design the caching and serving infrastructure for this system.

    medium~4 min

  4. 16.Design a security scanning orchestrator for GitLab Security that coordinates SAST, DAST, and dependency scans across thousands of projects. How do you handle scan results that take 30 minutes to complete?

    medium~5 min

  5. 17.You need to design a feature flag service that every GitLab.com request hits for A/B testing and gradual rollouts. The service must stay available even when its database is down. Walk me through your approach.

    hard~5 min

  6. 18.Design a deployment system that can safely roll out GitLab application updates to 50+ Kubernetes clusters across multiple regions while maintaining zero-downtime for users.

    hard~5 min

Leadership Questions (6)

  1. 19.Tell me about a time you helped team members outside your immediate group solve a DevOps or infrastructure problem. How did you approach stepping into their area?

    easy~4 min

  2. 20.Tell me about a time you identified that your team's processes weren't working efficiently and you drove changes to fix them. What was broken and how did you implement the improvements?

    easy~4 min

  3. 21.Describe a time you had to convince stakeholders to take an iterative approach rather than building a comprehensive solution upfront. What was the situation and how did you make the case?

    medium~5 min

  4. 22.Walk me through a time when you had to lead incident response or a major outage resolution entirely through asynchronous communication. How did you coordinate the team?

    medium~5 min

  5. 23.Describe a situation where you had to influence senior engineers or architects to adopt a different technical approach, but you didn't have formal authority over them. How did you build consensus?

    hard~5 min

  6. 24.Walk me through a time you had to make a technical decision that would impact multiple teams at GitLab scale, but you had incomplete information and couldn't wait for perfect data. How did you approach the decision?

    hard~5 min

Problem Solving Questions (6)

  1. 25.GitLab.com serves 20 million registered users but we've noticed merge request load times have increased by 200ms over the past month with no obvious code changes. How would you estimate the business impact of this performance degradation?

    easy~3 min

  2. 26.You discover that GitLab.com's Redis clusters are running at 80% memory utilization during peak hours, but adding more Redis instances would cost $50K monthly. How would you estimate the risk and cost of not scaling versus the infrastructure investment?

    easy~4 min

  3. 27.You need to estimate how much additional infrastructure cost GitLab would incur if we enabled GitLab Duo's AI code suggestions for all free tier users instead of just paid users. Walk me through your calculation.

    medium~4 min

  4. 28.GitLab's CI/CD Auto DevOps feature automatically sets up pipelines for new projects, but we've seen a 15% increase in failed pipeline runs over the past quarter. No code changes were made to Auto DevOps itself. How would you systematically investigate what's causing this increase?

    medium~5 min

  5. 29.GitLab is considering offering dedicated CI runners as a premium feature. Estimate how many dedicated runners we'd need to provision globally to serve 10,000 enterprise customers while maintaining our current SLA commitments.

    hard~5 min

  6. 30.A security researcher reports that GitLab's dependency scanning misses 30% of known vulnerabilities compared to competitors. Our product team wants to improve accuracy but estimates it would double scanning time. How would you approach quantifying whether this trade-off is worth making?

    hard~5 min

More GitLab interview questions