Fractional DevOps Engineer
Senior ownership of your AWS infrastructure, 10–20 hours a week, on a monthly retainer.
Most of the startups that contact me have the same staffing problem. The infrastructure is real now: paying users, an EKS cluster someone set up in a hurry, a security questionnaire from the biggest customer, an AWS bill nobody has examined in months. There is clearly senior DevOps work to do.
But the work comes in bursts. One week needs twenty focused hours and the next needs three. That is hard to justify as a full-time senior salary, and it is too much to leave with a backend developer who has a product to build. A fractional arrangement fits this shape of work: I take ownership of the infrastructure for an agreed number of hours each week, and you pay for those hours. It is one of the engagement models I offer and the one most teams end up on.
What is a fractional DevOps engineer?
A fractional DevOps engineer is a senior DevOps engineer who owns your infrastructure part-time on a retainer, typically 10–20 hours per week, instead of joining as a full-time hire. The responsibilities are the same ones an in-house hire would carry: infrastructure as code, CI/CD, cluster operations, monitoring, security, incident response. The hours and the employment model are what differ.
Who this is for
The model fits startups and small product teams on AWS that have crossed into real infrastructure but cannot fill a forty-hour week with senior DevOps work. Concretely:
- Production users who notice when things break.
- Compliance pressure starting to arrive, whether that is customer security questionnaires or an upcoming SOC 2 audit.
- Scaling pains that show up at inconvenient times, usually in whatever part of the stack nobody owns.
- An AWS bill that grows a little every month and never gets reviewed.
This doesn't fit every team. If your product needs someone awake and responding at 3 a.m. every night of the year, a fraction of one engineer cannot provide that; you need an in-house team running a proper on-call rotation. I respond to incidents within the hours we agree on, and if your situation needs more than that, I will say so on the discovery call.
What I take ownership of
The retainer covers the work a senior infrastructure hire would own. Each area below is grounded in projects I have written up in detail.
Terraform and infrastructure as code
Your environments live in version-controlled Terraform with a plan-review-apply workflow, so every change is reviewed before it touches AWS. I wrote up how I structure this in the Terraform multi-environment AWS case study.
CI/CD pipelines
GitHub Actions or GitLab CI pipelines for building, testing, and deploying, with environment promotion and rollback paths your developers can use without asking me first.
EKS and ECS operations
Cluster upgrades, autoscaling, Helm packaging, and GitOps with ArgoCD. The setup I run is documented in how I run GitOps on a production EKS cluster with ArgoCD.
Monitoring and alerting
Prometheus and Grafana or CloudWatch, tuned so that an alert firing actually means someone should look. Dashboards your team reads, and runbooks for the alerts that page.
Security hardening
Secret scanning in pipelines, IaC compliance gates, IAM least-privilege reviews against actual CloudTrail usage, and dependency scanning. This comes from the organisation-wide rollout described in the DevSecOps pipeline transformation case study.
Incident response within agreed hours
When something breaks inside the response windows we have set, I am in the incident: diagnosing, fixing, and writing the postmortem afterwards.
Cost reviews
Regular passes through the bill to catch idle resources, oversized instances, and quiet month-over-month growth. The method is the one from my multi-account AWS cost optimization work.
How the engagement works
- Free discovery call. We walk through your infrastructure and where it hurts.
- Written proposal. Scope, hours, and what I would tackle first, in writing, so you can compare it against hiring or going with an agency.
- Fixed-scope start. Usually an audit or a contained project. You see how I work before committing to anything ongoing.
- Monthly retainer. 10–20 hours per week of ongoing ownership, adjusted as your needs change.
I do not publish retainer pricing because the answer depends on scope. A ten-hour week focused on cost reviews prices differently from twenty hours that include incident response windows. We scope it on the discovery call and you get the number in the written proposal.
Who you would be working with
I'm Manip Poudel. My day job is DevOps Team Lead at a Tokyo-based software company, where I run 10+ production AWS environments. I hold the AWS Certified Solutions Architect and HashiCorp Terraform Associate certifications, and I am based in Nepal working remotely with teams worldwide. More background is on the about page.
Frequently Asked Questions
What is a fractional DevOps engineer?
A fractional DevOps engineer is a senior engineer who owns your infrastructure part-time on a retainer, typically 10–20 hours per week. The difference from hiring a freelancer for a one-off project is continuity: the same person carries your stack, its incident history, and the backlog from month to month, the way an in-house hire would.
How many hours per week do I get?
Most retainers run 10–20 hours per week. We set the number during scoping based on what your infrastructure actually needs, and we can adjust it month to month as the workload changes.
What's the difference between fractional DevOps and a DevOps agency?
With an agency you typically get a team with rotating staff, and the person who handled last month's incident may have moved to another account. With a fractional engineer you get one named senior person who knows your stack and its history. I am in every call and every incident myself.
Do you work across time zones?
Yes. Nepal time overlaps European mornings and US evenings, and my current team is in Tokyo, so cross-time-zone work is my normal day. I work async-first, with scheduled overlap windows agreed at the start of the engagement.
How do we start?
Book a free discovery call through the contact form. After the call I send a written proposal with scope and pricing, and most engagements begin with a fixed-scope audit or project before moving to a monthly retainer.
Book a free discovery call
If you can share an architecture diagram, a Terraform repo, or last month's AWS bill ahead of time, I read it before we talk, so the first conversation starts from your actual setup.
Get In Touch