steve okoth / ~/portfolio

print  ·  Ctrl + P  or  Cmd + P  ·  save as pdf

Steve Okoth

Cloud & Platform Engineer

Full CV with email and phone available on direct request via GitHub or LinkedIn.

summary

Cloud Infrastructure Engineer with ten years across software engineering, DevOps, and cloud platform work. AWS Certified. Track record cutting cloud spend, accelerating deployment cadence, and tightening incident response — $50K/month saved, deploy frequency from 5 to 100/day, MTTR from 5 hours to 15 minutes. Comfortable across AWS, GCP, and Azure; Terraform, Ansible, Kubernetes, ArgoCD.

experience

  • Cloud Infrastructure Engineer II 2025-03 — present

    Tala · Remote

    • Lead automation of infrastructure and platform services across AWS, GCP, and Azure with Terraform and Ansible.
    • Architect new infra and platform solutions; deliver alongside development teams.
    • Design and run continuous deployment on Kubernetes via ArgoCD and Jenkins.
    • Application-performance alerting in Datadog and ElasticSearch — catch issues before they reach end users.
    • On-call rotation; incident response and post-incident hardening.
  • Cloud Support Engineer — Containers 2024-05 — 2025-03

    Amazon Web Services

    • Resolved customer cases for AWS container services across phone, email, and live chat.
    • Advanced troubleshooting to dive into root cause on complex container / orchestration issues.
    • Partnered with development teams on complex escalations and contact-deflection initiatives.
    • Wrote internal tools / scripts to streamline team workflows; contributed to process and policy compliance.
  • Senior DevOps Engineer 2021-10 — 2024-05

    Cellulant Kenya

    • Lifted deployment frequency from 5/day to ~100/day by automating the release pipeline in GitLab CI.
    • Cut lead time for changes from 2 days to 10 minutes by automating the change-approval flow.
    • Reduced change failure rate by 25% with rigorous unit / integration / performance gates before prod.
    • Cut MTTR from 5 hours to under 15 minutes — service backups, proactive monitoring, automated failover.
    • Reduced monthly cloud spend from $175K to $125K via right-sizing, reserved instances, and Fargate for spiky workloads.
    • Built an internal developer platform on Backstage — bootstrapping scripts, infra catalog, golden paths.
    • Automated all infrastructure provisioning in Terraform + Ansible.
  • Product Support Technical Lead 2021-05 — 2021-10

    Finserve Africa

    • Improved system uptime by 25% via better monitoring and proactive incident response.
    • Led the go-live of Jenga API and Jenga Payment Gateway integrations.
    • Streamlined support process — cut issue resolution time by 40%.
    • Enforced rigorous pre-release testing across product surfaces.
  • Integration Specialist 2020-09 — 2021-04

    NCBA Bank Kenya

    • Hit 98% SLA adherence across service-lifecycle management.
    • Cut integration errors by 30% and improved API response times by 25%.
    • Integrated four key systems; cross-functional response time up 35%.
    • Hardened security posture — 15% drop in security-related incidents.
  • HUB SIP Engineer 2018-12 — 2020-09

    Cellulant Kenya

    • Ran disaster-recovery drills; validated system resilience.
    • Cut application incidents by 20% via proactive monitoring.
    • New deployment strategy reduced system downtime by 15%.
    • Documentation overhaul — 25% drop in time-to-troubleshoot.
  • Software Engineer 2018-03 — 2018-11

    Sanergy

    • Streamlined the software release process — 20% reduction in errors.
    • Code-optimization measures improved system performance by 15%.
    • Led a team on a new module — 30% lift in user satisfaction.
  • Software Engineer 2016-08 — 2017-11

    Stanbic Bank Kenya

    • Contributed to the Payments and Collections initiative that materially grew the bank’s liabilities within the year.
    • Built the Receipt Referencing System — direct integrations with schools and businesses; drove deposit growth.
    • Integrated the bank’s tea-trading system to Equity Bank.
    • Customer risk-rating tool to score risk on loan and account products.
    • Self-service insurance-renewal flow for staff.

selected projects

  • A self-hosted, OCI-compliant container registry 2026

    Harbor and ECR's feature set — signing, scanning, SSO, audit — self-hosted, no cloud bill. Single-tenant by default; multi-tenant when you need it.

    Go · gRPC + mTLS · PostgreSQL · RabbitMQ · Helm · Cosign · OpenTelemetry

    75/75 OCI v1.1 conformance · 13 services · mTLS between every hop

  • Building Cellulant's Internal Developer Platform 2022 — 2024

    Engineers shouldn't need to know your stack to ship safely. So I built the platform that hides it.

    Backstage · GitLab CI/CD · ArgoCD · Kubernetes · Terraform · Ansible · Prometheus · Grafana

    5 → 100 deploys/day · 2 days → 10 min lead time · $50K/mo cloud saved

  • Cutting Cellulant's cloud spend without slowing the org 2023

    The AWS bill wasn't a tooling problem. It was a visibility problem.

    AWS Cost & Usage Reports · Athena · Reserved Instances · Fargate · VPC Endpoints · Backstage

    $175K → $125K / mo · 28% AWS bill cut · zero SLO regression

  • Janus — a self-hosted secrets manager 2026

    Doppler's ergonomics, Vault's crypto, your server. One Go binary, one Postgres, keys that never leave in plaintext.

    AES-256-GCM · AWS KMS · Go · PostgreSQL · Shamir SSS

    3-level envelope encryption · Shamir 3-of-5 + KMS auto-unseal · 100% crypto coverage

cloud certifications

  • Cloud Practitioner earned 2022-01 · expired 2025-01

stack

cloud
AWS (EKS, S3, EC2, VPC, IAM, NLB, ALB) · Azure · GCP
iac
Ansible · Pulumi · Terraform
containers
ArgoCD · Docker · Docker Swarm · Helm · Kubernetes
observability
Datadog · ElasticSearch · Grafana · New Relic · Prometheus
ci / cd
ArgoCD · Git · GitHub Actions · GitLab CI/CD · Jenkins
languages
Bash · Core Java · Go · PHP · Python · Spring Boot
data
MySQL · Oracle · PostgreSQL · RabbitMQ · Redis