DevOps Engineer

Work on real production platforms for UK engineering teams, remotely from home in South-East Asia. You join the client’s team as one of their engineers, and Thoughtgears contracts and pays you.

About Thoughtgears

Thoughtgears is a small UK company, run by its two partners, that places vetted DevOps and platform engineers with UK companies. We look for engineers who can be trusted with production, check that with one structured interview, and stay your point of contact for as long as you work with us.

What you will do

Own the infrastructure as code for a client’s platform, usually Terraform on GCP or AWS

Run and improve workloads on Kubernetes, and choose serverless where it is the better fit

Build CI/CD pipelines that make deploying safe and routine

Investigate production problems from logs, metrics and traces, fix them and help stop them coming back

Work in the client’s own channels, stand-ups and reviews, and explain your decisions in English

What you get

Real engineering on production systems, as part of the client’s team rather than a ticket queue

Fully remote from your home country, with no relocation

A fair, structured interview scored against a written rubric. No algorithm puzzles and no trick questions.

You do not pick a level: the interview sets it. After that we put you forward for roles that fit your level and stack, without another technical test from us.

Paid monthly in your local currency. Thoughtgears pays the transfer fees, so you receive the full amount.

You must have

Hands-on infrastructure as code (Terraform, Pulumi or CDK): you write and change resources yourself and understand plan, apply and state

Hands-on Kubernetes: you have deployed workloads to a cluster with kubectl or Helm, and know what Pods, Deployments and Services are for

You read logs, metrics and dashboards to find out what is going wrong

Clear spoken and written English for technical discussions with a client team

Ownership: you explain what you did and why, and raise problems early

Nice to have

Depth in GCP or AWS beyond the core services

Progressive or canary rollouts and deploy gates

Least-privilege IAM, workload identity and image scanning

An eye for cloud cost

What we expect at junior level

Does the everyday work on their own: Terraform changes, deployments to an existing cluster, pipeline changes

Learns design and production decisions from the senior engineers around them

What we expect at Mid-Level

Works independently on common production work: reusable modules, state and environments

Writes Deployments with sensible probes, requests and limits, and debugs failing ones

Builds a pipeline end to end, with caching, artifacts and promotion between environments

What we expect at Senior-Level

Production-grade on a client’s system without supervision

Structures infrastructure as code for a team and reviews others’ plans for risk

Designs rollouts, autoscaling and health checks around real readiness

Runs an incident calmly: mitigates first, finds the causes, proposes prevention

Explains trade-offs to engineers and non-engineers alike

What we expect at Lead-Level

Designs the platform and sets the standard others follow

Sets the infrastructure-as-code strategy and promotion between environments

Builds the reliability practice: SLOs, alerting and blameless incident reviews

Leads technical conversations with client stakeholders and mentors other engineers

Where and when

Remote, from South-East Asia (for example Indonesia, Malaysia, the Philippines or Vietnam). Your working day overlaps with the client’s UK working hours, so it usually starts in the afternoon and runs into the evening, your time. The exact hours are agreed with each client team. This is a contract role.

How we vet

You send us your CV

We review it. If it is a fit, you get an email with a login to our candidate portal, where you complete a short work-style questionnaire and record an introduction video in English

If we would like to talk, we send you a calendar invitation for an intro call

You take a technical interview of about an hour: talk through a system you have owned, diagnose a realistic production fault on your main cloud (GCP or AWS), and fix a small Dockerfile and pipeline on a shared screen

If you pass, you join our bench and we put you forward for roles that fit your level and stack

If you need a reasonable adjustment for any step, such as extra time or camera off, tell us at any point.

If you have done this work in production, we would like to hear from you, even if you do not match every point above.