What is DevOps?
DevOps is a set of practices that shortens the path from code change to running software by removing the separation between development and operations. It is organizational and cultural before it is technical.
#What DevOps changes
The traditional split had developers writing software and handing it to operations to run, which produced misaligned incentives: developers rewarded for shipping and operations rewarded for stability, with the handoff as the friction point.
DevOps merges the responsibility. Teams that build software also run it, which means they experience the operational consequences of their design choices directly. The technical practices, including automated deployment, infrastructure as code, and shared monitoring, exist to support that shared ownership rather than being the point in themselves.
#Why DevOps works
Feedback loops shorten in every direction when the same people build and operate.
- Aligned incentives: the team that gets paged designs for fewer pages.
- Faster delivery: no handoff queue between writing and shipping.
- Operational awareness: production behavior informs design decisions directly.
- Reduced knowledge gaps: the people who understand the code are the ones debugging it.
#Limits and common misapplications
The most common failure is treating DevOps as a job title or a tooling purchase. Renaming the operations team and adopting a pipeline tool without changing who is accountable for production leaves the original handoff intact with new vocabulary.
It also spreads operational expertise thinner. Deep skills in networking, storage, and security are hard to maintain in every product team, which is why most organizations run a platform team providing shared infrastructure while product teams own their services on top of it.
#DevOps vs SRE
DevOps is a broad philosophy about shared ownership and shortened feedback loops. It describes a direction more than a specification, which is why implementations vary widely.
SRE is a specific implementation with concrete mechanisms: SLOs, error budgets, toil budgets, and defined production readiness criteria. The often quoted framing is that SRE is one prescriptive way to do DevOps. In practice, DevOps tells you that developers should own production and SRE tells you exactly how to decide when to stop shipping features.
#Key takeaways
- DevOps merges build and run responsibility so teams experience the consequences of their design choices.
- The technical practices exist to support shared ownership rather than being the goal themselves.
- Renaming the ops team without changing accountability leaves the original handoff in place.
- SRE is a prescriptive implementation of DevOps with concrete mechanisms such as SLOs and error budgets.
#Frequently asked
Is DevOps a role?
It describes a way of organizing work rather than a job. Titles containing it usually refer to platform or infrastructure engineering.
Does every team need to run its own infrastructure?
No. Most organizations use a platform team that provides shared infrastructure while product teams own their own services on it.
How is DevOps adoption measured?
Through delivery outcomes rather than tool adoption. The DORA metrics are the standard measures.
Related terms
- 2 min
Site Reliability Engineering (SRE)
SRE conceptsSite reliability engineering is the discipline of applying software engineering methods to operations problems.
- 2 min
CI/CD
DeploymentCI/CD is continuous integration and continuous delivery or deployment: the automated pipeline that takes a code change from commit through build, test,…
- 3 min
DORA Metrics
Reliability & SLOsDORA metrics are five measures of software-delivery performance identified by the DevOps Research and Assessment program: change lead time, deployment…
- 2 min
On-Call
Incident managementOn-call is the rotation that guarantees a human is available to respond to production problems at any hour.