Senior Platform Engineer

Deogratias
Uwizeyimana

Running the Kubernetes platform behind 40+ banks.

I build and operate AKS-based managed hosting at Backbase, from GitOps and policy to observability and 24/7 incident response, so it stays secure, observable and boringly reliable.

View work ↓ Get in touch →
Portrait of Deogratias Uwizeyimana 
$ kubectl describe engineer deo
Role:      Senior Platform Engineer
Location:  Kraków, PL · remote EU
Status:     Open to new roles
Certs:     6/6 Ready
Uptime:    5y+ in cloud
40+
banks on the platform
I help operate
10+
production environments
migrated to AKS
50+
repos & pipelines moved
to GitHub Actions
24/7
on-call incident
response

Selected Work

Data services · Backbase · 2026 – Present

Multi-tenant Managed Redis & PostgreSQL re-architecture

Active–active
geo-replicated cache
Cache Azure Cache for Redis Managed Redis · region A Managed Redis · region B
Database Shared PostgreSQL server One server per application
Challenge

Client environments needed a path off Azure Cache for Redis. A single shared PostgreSQL server made it impossible to isolate applications from each other.

What I built
  • Designed and implemented a multi-tenant Azure Managed Redis platform with active-active geo-replication, so clients could migrate off Azure Cache for Redis.
  • Re-architected PostgreSQL provisioning into a multi-server model, migrating all existing state in place to give each application its own server.
Azure Managed Redis Azure PostgreSQL Terraform AKS Helm
Platform operations · Backbase · 2026 – Present

Keeping a 40-bank platform current, and leaner after hours

40+
banks · 24/7 on-call
Capacity
Off-hours ArgoCD SyncWindows Kyverno DeletingPolicy automatic scale-down
Challenge

Core components (AKS, Istio, ArgoCD and Kyverno) release fast, and banks expect zero surprises. Cluster workloads also ran at full scale outside working hours.

What I do
  • Own lifecycle upgrades of AKS, Istio, ArgoCD and Kyverno across the platform.
  • Automated off-hours workload scale-down with Kyverno DeletingPolicy and ArgoCD SyncWindows, improving resource use and consolidating tooling.
  • Take part in a 24/7 on-call rotation handling production incidents for 40+ banks.
  • Author SOPs that help support teams and internal developers adopt new platform features faster.
AKS Kyverno ArgoCD Istio Kubernetes Incident response
Observability · Backbase · 2023 – 2025

Datadog → Grafana: observability platform migration

4
alert channels, one framework
Telemetry Datadog PrometheusLoki Grafana
Alerting Grafana alerts PagerDutySlackTeamsEmail
Challenge

The Grafana stack offered more flexibility at lower cost, but high metric cardinality slowed dashboards down. Incident notifications also had no common standard across teams and customers.

What I built
  • Contributed to the strategic migration from Datadog to Grafana, improving flexibility and reducing costs.
  • Introduced adaptive metrics and caching for the Prometheus and Loki datasources, cutting cardinality and speeding up dashboard queries.
  • Built a configurable multi-channel alerting framework (PagerDuty, Slack, Microsoft Teams, email) to standardise incident notification.
Grafana Prometheus Loki Datadog PagerDuty Kubernetes
Cloud migration · Backbase · 2023 – 2025

AWS → Azure: consolidating client platforms on AKS

10+
prod envs migrated
Runtime AWS EKSPrivate Azure Managed Hosting · AKS
GitOps FluxCD ArgoCD + Helm
CI/CD Azure DevOps GitHub Actions50+ repos
Challenge

Production client environments were spread across AWS EKS and private Azure tenants. They had to move to a single managed hosting platform on AKS, for regulated banking workloads.

What I built
  • Migrated 10+ production client environments from AWS EKS and private Azure to Managed Hosting on AKS.
  • Standardised GitOps by moving cluster workload management from FluxCD to ArgoCD with Helm charts.
  • Migrated 50+ repositories and CI/CD pipelines from Azure DevOps to GitHub, adopting GitHub Actions.
  • Mentored junior engineers and ran internal knowledge-sharing sessions.
AKS ArgoCD Helm GitHub Actions FluxCD AWS EKS Terraform

Certifications

CKS
Certified Kubernetes Security Specialist
CNCF · verify ↗
CKA
Certified Kubernetes Administrator
CNCF · verify ↗
AWS
DevOps Engineer – Professional
Amazon Web Services · verify ↗
AWS
SysOps Administrator – Associate
Amazon Web Services · verify ↗
AZ
Azure Administrator Associate
Microsoft · verify ↗
TF
Terraform Associate
HashiCorp · verify ↗

Stack

Cloud
AzureAWS
Containers
KubernetesDockerHelmIstioKyverno
IaC & CI/CD
TerraformArgoCDGitHub ActionsFluxCD
Observability
GrafanaPrometheusLokiDatadogPagerDuty
Scripting
BashPython
AI tooling
Claude CodeOpenCodeMCP servers

About

I'm a platform engineer specialising in Kubernetes infrastructure, cloud migration and observability for production-critical, regulated environments. At Backbase I help run the AKS-based managed hosting platform that 40+ banks depend on.

My principle is reliability by design, not by rescue: security, operability and resilience are built in from the start, not bolted on after the first incident.

I came to cloud from engineering: a dual Master's across AGH Kraków and IST Lisboa, then IT support at HCL, before moving into cloud at Andela and Backbase. I'm fluent in English and French and used to working across international teams.

  1. Senior System EngineerHEAD
    Backbase · Platform Engineering · 2026 – now
  2. System Engineer
    Backbase · Migration & Observability · 2023 – 2025
  3. Junior System Engineer
    Backbase · EKS & AKS onboarding · 2021 – 2022
  4. Associate Cloud Engineerandela
    Andela · GCP, remote · 2021
  5. IT Analyst
    HCL Technologies · Kraków · 2020 – 2021
  6. Dual Master's degreeinit
    AGH Kraków / IST Lisboa · Chemical Technology & Energy Engineering · 2019 – 2021

Let's work
together.

Open to senior platform and infrastructure roles, remote across Poland and the EU.

[email protected] LinkedIn ↗