Career insights / SRE
Site Reliability Engineer
Treat reliability as an engineering problem you can measure.
Site Reliability Engineers apply software engineering to operations. They agree reliability targets with product teams, measure them, automate repetitive work and lead incident response, so users stay happy while new releases keep shipping.
Indicative ranges from public salary sources. Not offers or guarantees.
A day in the life
What the work actually looks like, hour by hour.
Check dashboards and overnight alerts. Review error-budget burn for the services you own.
Triage tickets and the on-call handoff. Fix noisy alerts so the pager only rings for real problems.
Join a launch review with developers: load-test results, rollback plan, dashboards and runbooks.
Automate toil: replace a manual weekly restart with a script, scheduled job or Kubernetes operator.
Write or review a blameless postmortem from last week's incident and track its action items.
When paged: acknowledge, assess user impact, mitigate first (roll back, scale, fail over), investigate after.
What you will own
Define SLIs and SLOs with product owners
Build dashboards and alerts across metrics, logs and traces
Lead incident response and blameless postmortems
Reduce toil with automation and self-healing systems
Plan capacity and run load tests
Design safe releases: canary, blue-green and automatic rollback
Run game days and chaos experiments
Keep runbooks current and on-call sustainable
Core concepts to master
SLI
A measurement of service health, such as the percentage of requests that succeed or finish under 300 ms.
SLO
The target for an SLI over a period, for example 99.9% of requests succeed over 30 days.
SLA
A customer contract that defines what happens, usually credits or penalties, if a target is missed.
Error budget
The unreliability the SLO allows. When it is spent, risky releases pause and reliability work comes first.
Toil
Manual, repetitive work that can be automated and grows with the service. SREs aim to keep it low.
Golden signals
Latency, traffic, errors and saturation: the four signals to watch on almost any service.
Observability
Understanding a system from its outputs: metrics for trends, logs for events, traces for request paths.
Postmortem
A blameless write-up of what happened, why, and which actions stop it recurring.
Error budget calculator
Pick an availability target to see how much downtime it allows.
Example incident, minute by minute
An illustrative scenario for learning, not a real customer incident.
Alert fires: error-budget burn is 6Γ normal on the checkout service.
On-call engineer acknowledges and checks the dashboard: error rate 4%, latency rising, a deploy went out 10 minutes ago.
Mitigate first: roll back to the previous version. Error rate drops back to normal.
Confirm recovery with the golden signals and tell stakeholders the impact has ended.
Blameless postmortem: a bad config change reached production. Actions: add a config check to the pipeline and a canary stage.
The loop in motion
$ ready
Toolbox
Metrics & alerting
Logs & traces
Platform
Automation
Learning roadmap
Follow these stages in order. Each one builds on the last.
- 1
Linux, networking and Git
The base every incident investigation starts from.
- 2
Docker and Kubernetes
Run and debug containerised services.
- 3
CI/CD and release strategies
Canary, blue-green, rollbacks.
- 4
Metrics and dashboards
Prometheus, PromQL and Grafana.
- 5
Logs and tracing
Loki or ELK, plus OpenTelemetry basics.
- 6
SLOs, alerts and error budgets
Turn reliability into numbers and actionable alerts.
- 7
Incident response and postmortems
Roles, communication, mitigation first.
- 8
Automation and chaos
Python tooling, game days and failure testing.
Capstone project
Kubernetes SRE Command Center: observe workloads, inspect events, query metrics and logs, and build an AI-assisted root-cause workflow.
Career path
Junior SRE / DevOps engineer
Learn the stack, handle tickets, join on-call as a shadow.
SRE
Own services, SLOs and alerts. Lead small incidents.
Senior SRE
Design reliability for whole systems. Mentor others.
Staff / Principal SRE
Set reliability strategy across teams.
SRE manager
Lead the team, on-call health and reliability programs.
Is it for you?
- You enjoy debugging and asking why a system failed.
- You are comfortable with Linux and scripting, or willing to practise daily.
- You stay calm under pressure and like writing things down clearly.
Interview questions to practise
What is the difference between SLI, SLO and SLA?
An SLI is a measurement, an SLO is the target for it, and an SLA is a contract with consequences if the target is missed.
What is an error budget?
The amount of failure the SLO allows. At 99.9% over 30 days that is about 43 minutes of downtime. When it is used up, risky releases pause and reliability work takes priority.
How do you handle a high-latency incident?
Confirm user impact, mitigate first (roll back, scale or shed load), then find the cause using the golden signals and recent changes. Keep stakeholders updated and write a blameless postmortem.
What are the four golden signals?
Latency, traffic, errors and saturation.
What is toil and how do you reduce it?
Manual, repetitive, automatable work with no lasting value. Reduce it by automating, fixing root causes and removing unnecessary steps.
How do you reduce alert fatigue?
Alert on symptoms tied to SLOs, remove duplicates, link every alert to a runbook, and review noisy pages weekly.
Common questions
Is SRE harder than DevOps?
It is different. SRE leans more on measurement, debugging and incident response. Strong DevOps skills are the best foundation.
Do SREs need to code?
Yes. Python or Go for automation and tooling is common, along with strong Bash and Linux.
Is on-call required?
Most SRE roles include on-call rotations. Good teams keep them fair and limit toil.
Train for SRE with a mentor
Hands-on labs, real projects, mock interviews and placement preparation at our Belagavi campus, or online.