The State of SRE/DevOps Interviews in 2025: A Reality Check

Reflections from a decade-long SRE/DevOps veteran after four months of intensive job searching

After over a decade working as a Production Engineer, SRE, and DevOps professional across companies ranging from startups to FAANGs, I recently found myself back in the job market. Over the past four months, I've interviewed with approximately 32 companies, going through 2 interview processes per week. This intensive experience has given me a unique window into the current state of SRE/DevOps hiring practices—and frankly, it's a mess.

The Paradox of Uniform Job Descriptions, Chaotic Interview Processes

Here's what struck me most: despite nearly identical job descriptions across companies, the interview processes are wildly inconsistent. Every posting mentions the same requirements:

• AWS cloud infrastructure experience

• Infrastructure management with open-source tools (Terraform, Kubernetes)

• Scripting experience in Python or Bash

• General "DevOps" responsibilities

Yet the interviews themselves range from reasonable technical discussions to completely misaligned assessments that seem designed for different roles entirely.

What Companies Are Actually Looking For

Through this gauntlet of interviews, I've identified what hiring managers really want (regardless of what their job descriptions say):

1. Strong Coding Skills: Not just scripting—they want candidates who can solve LeetCode medium-level algorithmic problems

2. Incident Management Expertise: Deep knowledge of on-call processes and structured troubleshooting methodologies

3. Systems Fundamentals: Solid understanding of operating systems and networking principles

4. Language Proficiency: Deep expertise in at least one programming language (Python and Golang are particularly in demand)

5. Code Quality Skills: Ability to debug existing code and refactor legacy systems

The Quant/Fintech Extreme

Quantitative trading firms and fintech companies have taken interview complexity to another level entirely. Their assessment process often includes:

• Aptitude Tests: Mental math and logical reasoning problems

• SQL Challenges: Complex database query optimization

• Coding Assessments: Mix of algorithmic problems and practical exercises (log parsing, financial data processing)

• Hardware Knowledge: Obscure questions about specific networking hardware brands (yes, Citadel actually asked an SRE candidate about Solarflare cards)

The Fundamental Disconnect

Here's the core problem: companies are making interviews unnecessarily complex while losing sight of what they're actually hiring for. If someone's resume shows relevant projects and experience, 50% of the evaluation should already be complete. The interview should focus on validating that experience, not testing unrelated algorithmic problem-solving skills.

I've witnessed interviewers ask graph traversal questions for SRE roles. This isn't just misguided—it's actively counterproductive.

The Experience Gap Problem

The shift to cloud-native infrastructure has created an interesting dynamic. Most candidates today have primarily cloud and Kubernetes experience. The engineers with deep traditional systems knowledge have largely moved into leadership roles or retired at recently successful companies. This leaves a gap where experienced engineers may not have the coding interview skills expected by top-tier companies, despite having the actual job-relevant experience.

Interviewer Quality Issues

A concerning trend I've observed is the quality of interviewers themselves. Multiple candidates (including myself) have encountered situations where:

• Interviewers seem unprepared or lack adequate knowledge to conduct technical assessments

• Some remain completely silent, offering no guidance or feedback during coding exercises

• Interviewers are surprised by standard solutions to problems they've assigned

In one Apple interview, a software engineer gave me a LeetCode hard problem, then seemed genuinely shocked when I solved it using recursion. This has happened twice in the past two years. When the people evaluating you don't understand your solutions, the entire process breaks down.

The Bright Spots

Not all companies handle this poorly. Meta, despite its other issues, has consistently delivered well-structured interviews. Their approach is clear: "Interview to Hire." They've figured out how to assess candidates fairly while maintaining high standards.

The AI Factor

The elephant in the room is artificial intelligence. Industry leaders are openly discussing how entry-level white-collar jobs will be significantly impacted by AI within the next five years. This raises serious questions about how newcomers will enter the SRE/DevOps field when both experience requirements and job availability are becoming more restrictive.

A Path Forward: Redefining Interview Processes

The solution isn't to eliminate technical interviews, but to align them with actual job requirements. Here's my proposed framework:

Step 1: Honest Job Titles and Project Expectations

Companies should clearly define not just the role title, but also the actual work candidates will be doing:

• Software Developer with Infrastructure Experience: Tool-building roles requiring heavy development

  • Projects you'll work on: Building internal CLI tools, infrastructure APIs, deployment utilities

• Support Engineer/Linux Engineer: System administration focused positions

  • Projects you'll work on: Server maintenance, performance tuning, troubleshooting production issues, user support, capacity planning

• Cloud Infrastructure Engineer: AWS/Azure-focused roles without heavy development

  • Projects you'll work on: Managing cloud resources, cost optimization, security compliance, infrastructure monitoring, vendor integrations

• Site Reliability Engineer: Traditional SRE work (SLOs, incident management, change management)

  • Projects you'll work on: Defining service level objectives, building alerting systems, post-mortem analysis, reliability improvements, chaos engineering

• Platform Engineer: Developer-focused roles building internal tools, platforms, and developer experience solutions—essentially software engineers with infrastructure domain expertise

  • Projects you'll work on: Building developer platforms, CI/CD systems, internal PaaS solutions, developer tooling, infrastructure abstractions

Step 2: Matched Interview Processes

For roles with ~25% coding:

• Code debugging exercises instead of algorithmic problems

• Structured troubleshooting scenarios

• Incident management case studies

• CI/CD implementation discussions

• Systems knowledge assessment (Linux, networking, OS fundamentals)

For roles with ~50% coding:

• Code refactoring exercises (give poorly written code, ask for improvements)

• Infrastructure-focused system design (monitoring systems, deployment pipelines, alerting architectures)

• Technical deep-dives into past projects

For roles with 80%+ coding:

• Follow traditional software engineering interview processes

The Bottom Line

The current state of SRE/DevOps interviews is broken. Companies are using misaligned assessment methods that don't predict job performance while creating unnecessary barriers for experienced professionals. The solution requires honest self-assessment from hiring managers about what they actually need and structured interview processes that evaluate those specific skills.

For candidates navigating this landscape: prepare for everything, expect inconsistency, and don't take rejections personally when the process itself is fundamentally flawed.

What has your experience been with SRE/DevOps interviews? Share your stories in the comments—the community needs to hear about these challenges to drive change.