Platform engineer for systems that need to hold.
I build the platforms, automation and operating practices that help service teams own their reliability — from telemetry and alerting to the delivery paths behind them.
Miles Whittaker · Observability engineer · UK → Germany
I make complex systems
easier to run.
I lead observability and incident-management change across hybrid estates, turning monitoring, alerting and delivery into reliable, team-owned systems.
50+ teams supported
3,000+ services in the estate
800+ PagerDuty users
MW / 01
About / 01
Reliability is a team
sport.
I’m Miles, an observability engineer at bet365. I came up through operations and infrastructure, so I care about the work after a system is shipped: how people find the signal, make a decision, and recover when it matters.
Today I help teams move from centralised monitoring and triage to clear ownership, automated delivery and useful observability. The tools vary — New Relic, PagerDuty, Terraform, Ansible, GitLab and GCP — but the aim does not: fewer blind spots, less toil and a faster route to the right action.
Experience / 02
The useful
30 seconds.
Observability engineer specialising in large-scale monitoring and incident-management transformation across hybrid environments — on-prem and GCP.
Credentials that travel.
- CKACertified Kubernetes Administrator · Linux Foundation · Aug 2026—Aug 2028
- GCP ACEAssociate Cloud Engineer · Google · Jul 2024—Jul 2027
- AWS SAAAWS Solutions Architect—Associate · previously held · expired Oct 2025
One story, ready to send.
Open to Platform, SRE, Observability and Cloud Engineering roles in Germany, with Berlin a primary target. The CV and portfolio are designed to use the same language.
Request the CV ↗Selected experience
Problem → intervention → result.
Jan 2024—PresentObservability Engineer · bet365
Led observability and incident-management transformation across 50+ teams, 800+ PagerDuty users and 3,000+ services.
Apr 2023—Jan 2024IT Operations Engineer · bet365
Owned monitoring, incident triage, service continuity and change operations for mission-critical systems.
DE
Open to work
Berlin, Germany + other locations · hybrid or remote.
Based in the United Kingdom and actively exploring Germany-based platform and observability opportunities. English is the authoritative site language for now.
Selected engineering work / 03
Proof, not promises.
Each project will become a case study with the same honest shape: the problem, the constraint, the decision, and what changed.
Platform Atlas
A living map of a multi-site platform: delivery, ingress, observability and recovery, translated into something a new teammate can actually navigate.
Enter the system ↘signal / received
Observability transformation
New Relic, PagerDuty and Telegraf across 50+ teams and 3,000+ services — with automation making monitoring consistent, version-controlled and self-service.
Read the case study ↗planapplyverify
IaC & delivery systems
Terraform, Ansible, GitLab CI/CD and GCP Cloud Deploy combine into a delivery path that can be linted, tested, reviewed and repeated.
Read the case study ↗Interactive proof / 04
Platform Atlas.
A public-safe map of the platform thinking behind the work: delivery, ingress, observability and recovery, translated into something a new teammate can navigate.
Sanitised demo dataDrag-free by design · click a flow
Evidence lab / 05
One release.
Three signals.
A small, public-safe fixture showing how the same deployment can be investigated through profiles, traces and a change timeline.
Continuous profiling / Go
Where did the CPU move?
baseliner2026.09.12
http.Server100%
serveHTTP61%
json.Marshal22%
encodeValue9%
candidater2026.09.13
http.Server100%
serveHTTP72%
template.Execute31%
escapeString14%
previous deploycurrent deploysample: 60 s CPU profile
OpenTelemetry / Tempo
Which span explains the wait?
GET /checkout
184 ms
auth middleware
18 ms
catalog service
74 ms
postgres query
112 ms
render response
21 ms
0 ms50100150200 ms
span durationinvestigate firsttrace id: 7f3a…c91e
Release evidence / Git
What changed between runs?
r2026.09.12 baseline
Profile capturedsteady-state traffic · 60 s sampler2026.09.13 candidate
Template path changedprofile diff → trace check → promote01build02profile03trace04decision
candidate releasecorrelation is a lead, not prooffixture data only
Synthetic, redacted evidence for demonstration. It shows an investigation shape; it does not claim to measure a live service or expose KNR telemetry.
How I work / 06
Clarity is a feature.
The most useful systems are not just resilient. They explain themselves when the pressure is on.
Design for the handover.
Good work survives the person who made it. Interfaces, runbooks and architecture should leave the next decision easier than the last.
Make evidence cheap.
Observability is not a wall of charts. It is the shortest path from “something changed” to “we know why”.
Keep the edges honest.
Public surfaces get the polish. Private systems get the boundaries. A demo should be impressive without pretending to be live.
Ship the smallest proof.
Start with a useful, measurable slice. Let the shape of the real problem earn the next layer of complexity.
Open channel / 07
Have a system
worth making clearer?
For platform work, reliability engineering or a Germany-based opportunity, I’d like to hear what you’re building.
Start a conversationhello@alias.whitt.ukManchester · UK → Germany