StockFlow — Multi-Tenant Inventory & Order Management
B2B wholesaler back office on an append-only stock ledger, with the AWS infrastructure to run it.
- NestJS
- Next.js
- PostgreSQL
- Redis
- ECS Fargate
- +6
Available for DevOps & platform roles
DevOps Engineer
I build the pipelines and infrastructure that put code in production safely.
whoami --verbose
01/About
I run production infrastructure in the cloud: Kubernetes workloads across managed and self-managed clusters, Helm charts in place of hand-written YAML, and ArgoCD so the cluster matches Git rather than whatever someone applied by hand. Terraform provisions it and GitHub Actions ships it.
Security is part of that job, not a separate one. Snyk and Trivy gate the pipeline, secrets come from Vault, CI authenticates over OIDC with no stored keys. It also has to survive an audit, so I have worked with Delve on compliance automation and on the infrastructure side of SOC 2 and GDPR.
What I optimise for is the boring kind of reliability. Credentials that expire on their own, pipelines that fail loudly before production does, and infrastructure someone else can read six months later without calling me.
permissions:
id-token: write
contents: read
- uses: configure-aws-credentials@v4
with:
role-to-assume: ${{ secrets.ROLE_ARN }}
aws-region: eu-north-1
# no AWS_SECRET_ACCESS_KEY anywhereHow the pipelines authenticate to AWS, with no stored keys.
02/Skills
03/Experience
Dec 2025 - Present
Production infrastructure for client projects: cloud, CI/CD, security scanning and secrets management.
Jun 2025 - Nov 2025
CI/CD automation, containerization and AWS infrastructure for live projects.
2024 - 2025
Built and operated real infrastructure independently to develop depth across cloud, Kubernetes and automation.
2023 - 2024
Full-stack web applications with payments, authentication and cloud deployment.
04/Projects
Client work sits alongside self-directed builds. Where a repository is private, it says so rather than linking somewhere unhelpful.
B2B wholesaler back office on an append-only stock ledger, with the AWS infrastructure to run it.
Full-stack ecommerce on a custom VPC, with no static AWS keys and no open SSH.
Client production platform on ECS Fargate, with VPN-only access to internal services.
MERN rental marketplace with Stripe payments and an admin panel. Final year project.
05/Certifications
Amazon Web Services
Amazon Web Services
Course: CentOS & Ubuntu administration, shell scripting, user and permission management
School of DevOps (Udemy): playbooks, roles, loops, server automation
Course: MongoDB, Express, React, Node.js
06/FAQ
Direct answers to what recruiters and engineers ask most often.
Terraform for provisioning infrastructure in production, Ansible for server provisioning and configuration management across multiple hosts, and Helm for Kubernetes releases. Pipeline and cluster configuration is version-controlled the same way application code is.
Yes. Alongside AWS he has worked with Hetzner Cloud, provisioning servers, running a self-managed k3s Kubernetes cluster on them, and using Hetzner S3-compatible object storage, and with RunPod for GPU infrastructure to deploy models. The tooling around them (Terraform, Ansible, containers, GitOps) transfers across providers.
Supabase and PostgreSQL, including Row Level Security policies for access control at the database layer rather than only in application code. Also AWS RDS in production, where he configured automated backups and resolved security group misconfigurations, and MongoDB for Node.js applications.
07/Field Notes
Write-ups from production work.
$ cat ingressnightmare-cve-2025-1974.md
KubernetesA critical unauthenticated RCE in the ingress-nginx admission webhook, and how we closed it on a production Kubernetes cluster without dropping traffic.
5 min read
Read →08/Contact
Open to DevOps, platform and cloud infrastructure roles. If you have a pipeline that keeps breaking or infrastructure that needs a second pair of eyes, send a note.