Skip to content
HI
Available for new engagementsAntananarivo, Madagascar · Remote worldwide (UTC+3)

I build cloud platforms that ship faster.

Cloud & Platform Architect

6+ years designing, industrializing and operating mission-critical Kubernetes and OpenShift platforms on Azure, AWS and hybrid cloud — so engineering teams deploy in minutes, not weeks.

Start a conversation Explore my work
  • Kubernetes & OpenShift in production
  • Azure · AWS · Hybrid
  • Everything as Code
Hercule Ihasa

Hercule Ihasa

Cloud & Platform Architect

About

Platforms are products. I build them like one.

I'm a Cloud & Platform Architect and DevSecOps Lead who turns fragile, hand-operated infrastructure into governed, self-service platforms. Today I lead the DevSecOps practice at IDWEB / AMWEB (Groupe Constellation), where I own the technical strategy for the group's Kubernetes and OpenShift estate across France, Mauritius and Madagascar.

Mission

Give every engineering team a paved road to production — automated, secure by design, observable end to end.

Vision

Infrastructure should be invisible. When the platform is right, developers stop thinking about it and start shipping value.

One-line philosophy

“Automate everything, secure it from day zero, and build platforms that accelerate business value.”

Automation First

If a human did it twice, it becomes code. Toil is a bug.

Security by Design

Controls belong in the pipeline and the platform — not in a PDF nobody reads.

Observability Everywhere

You can't operate what you can't see. Metrics, logs and traces are non-negotiable.

Radical Ownership

I run what I build. Production is the only environment that tells the truth.

By the Numbers

Engineering impact, quantified

0+

Years in cloud & DevOps

From systems administration to platform leadership

0+

Production clusters governed

Kubernetes & OpenShift across Azure, AWS and hybrid

0

CI/CD ecosystems industrialized

GitLab CI, GitHub Actions, Jenkins, Azure DevOps

0%

Infrastructure as Code

Terraform + Ansible + GitOps — zero click-ops on the paved road

0

Countries in daily collaboration

Madagascar, France, Mauritius

0

Observability platforms deployed

Prometheus, Grafana, ELK, Loki, Instana

Figures reflect scope across my platform and DevSecOps work at Groupe Constellation and prior engagements.

Architecture Highlights

The four systems I master end to end

Not a list of logos — the actual operating model I design, industrialize and run in production.

Kubernetes / OpenShift

Enterprise Container Platforms

Multi-tenant Kubernetes and OpenShift clusters designed for isolation, quota governance and day-2 operations at scale.

  • Multi-tenant KaaS with namespace-as-a-service onboarding
  • OpenShift production operations & lifecycle (MCO)
  • Helm-based golden paths for application teams
  • Capacity, upgrade and disaster-recovery playbooks
CI/CD & GitOps

Delivery Pipelines as a Product

Standardized pipeline templates rolled out group-wide — one paved road from commit to production, whatever the team.

  • GitLab CI, GitHub Actions, Jenkins, Azure DevOps
  • GitOps-driven promotion across dev / staging / prod
  • Reusable pipeline libraries with built-in quality gates
  • Deployment frequency up, lead time and rollbacks down
DevSecOps

Security Wired Into the Pipeline

Shift-left security that developers don't have to think about: scanning, policy and secrets handled by the platform.

  • SAST / DAST / IAST integrated into every pipeline
  • Container image scanning and admission policies
  • Centralized secrets management
  • Compliance and governance mapped to controls-as-code
Observability

Full-Stack Visibility

A complete telemetry stack so incidents are detected by dashboards — not by customers.

  • Prometheus & Grafana for metrics and alerting
  • ELK and Loki for centralized logging
  • Instana APM for distributed tracing
  • SLO-driven alerting and on-call ergonomics

Career Journey

From security intern to platform leader

A deliberate progression: systems → cloud → platforms → leading the people who build them.

  1. Jan 2025 — PresentCurrentPromotion

    Team Leader DevSecOps · Cloud Platform Architect

    IDWEB / AMWEB — Groupe Constellation

    Promoted to own the group's cloud and platform technical strategy. I lead the DevOps team, set the Kubernetes/OpenShift architecture standards, and drive DevSecOps maturity across international entities.

    • Governance of production Kubernetes & OpenShift clusters for the whole group
    • Group-wide standardization of CI/CD, security and observability practices
    • Technical leadership of the DevOps team and its roadmap
    • Collaboration with teams and partners across France and Mauritius
  2. Dec 2021 — Dec 2024

    DevSecOps Architect · Cloud Platform Engineer

    IDWEB / AMWEB — Groupe Constellation

    Designed and operated the group's Kubernetes and OpenShift platforms on Azure and AWS — from architecture to 24/7 production operations.

    • Built a multi-tenant Kubernetes-as-a-Service platform for client environments
    • Industrialized CI/CD across GitLab CI, Jenkins, GitHub Actions and Azure DevOps
    • Automated the full infrastructure lifecycle with Terraform, Ansible and GitOps
    • Deployed the group observability stack: Prometheus, Grafana, ELK, Loki, Instana
  3. Feb 2021 — Nov 2021

    DevOps Engineer

    Hairun Technology

    Modernized an existing AWS platform: industrial CI/CD, Terraform-managed infrastructure and clean dev / staging / prod separation.

    • Migrated hand-managed AWS infrastructure to Terraform
    • Set up monitoring and observability from scratch
    • Cut delivery friction with standardized pipelines
  4. Jul 2019 — Feb 2021

    DevOps Consultant

    IT Consulting

    Advised clients on Kubernetes architecture, CI/CD automation and cloud deployment strategy.

    • Kubernetes architecture design for client workloads
    • CI/CD automation and delivery pipeline optimization
    • Infrastructure as Code adoption
  5. 2016 — 2018

    Early Foundations — Security & Systems Internships

    Orange Madagascar · École Nationale d'Informatique

    Information security operations at Orange Madagascar; systems and network administration at ENI. The security-first reflex started here.

    • Information systems security activities (Orange Madagascar)
    • Systems & network administration, infrastructure support (ENI)

Case Studies

Deep dives, not bullet points

Four flagship builds, each told the way engineering work should be judged: context, role, architecture, measurable impact.

Platform Engineering01

Kubernetes as a Service — a multi-tenant platform for client environments

Context & Challenge

Every new client environment meant weeks of manual cluster setup, inconsistent configurations and unpredictable security posture. The group needed to onboard client workloads fast without multiplying operational risk.

My Role

Platform architect and technical lead — I owned the design, the automation strategy and the production operating model.

Architectural Solution

  • Multi-tenant Kubernetes architecture with strict namespace isolation, quotas and network policies
  • Self-service onboarding: a new tenant environment is provisioned from templates, not tickets
  • Terraform + Ansible + GitOps as the single source of truth for cluster state
  • Golden-path Helm charts so application teams deploy without touching cluster internals

Impact

Weeks → hours

environment provisioning time

1 paved road

standard onboarding for every tenant

Drift ≈ 0

cluster state fully reconciled from Git

DevSecOps02

Security wired into every pipeline — SAST, DAST, IAST at group scale

Context & Challenge

Security reviews happened late, manually and inconsistently — a bottleneck for delivery and a blind spot for risk. The group needed security that scales with the number of teams, not with the number of security engineers.

My Role

DevSecOps architect — I defined the security-as-code strategy and drove adoption across development teams.

Architectural Solution

  • SAST, DAST and IAST stages embedded in the standard CI/CD templates
  • Container image scanning with enforced policies before any production admission
  • Centralized secrets management replacing credentials scattered in code and CI variables
  • Security findings surfaced in the developer workflow instead of end-of-cycle reports

Impact

100%

of standard pipelines run security gates by default

Shift-left

vulnerabilities caught at commit time, not in production

0 secrets

hardcoded in repositories on the paved road

Observability03

From blind operations to SLO-driven visibility

Context & Challenge

Incidents were detected by users before operators. Logs lived on individual machines, metrics were partial, and every diagnosis started from zero.

My Role

Architecture and rollout of the complete observability stack across platforms and applications.

Architectural Solution

  • Prometheus + Grafana for metrics, dashboards and alerting across all clusters
  • Centralized logging with ELK and Loki — one query surface for every environment
  • Instana APM for distributed tracing on critical applications
  • Alerting redesigned around symptoms and SLOs instead of raw host metrics

Impact

Minutes

to diagnose what previously took hours

1 pane

of glass for metrics, logs and traces

Proactive

dashboards catch degradation before users do

Infrastructure as Code04

Everything as Code — industrializing the entire infrastructure lifecycle

Context & Challenge

Manual infrastructure changes made every environment a snowflake: unreproducible, undocumented and risky to touch. Multi-environment consistency across Azure, AWS and hybrid was impossible by hand.

My Role

Automation lead — I drove the Terraform + Ansible + GitOps strategy from first module to full adoption.

Architectural Solution

  • Terraform modules as reusable building blocks for Azure and AWS landing zones
  • Ansible for configuration management and OS-level compliance
  • GitOps workflows: infrastructure changes ship through pull requests with review and history
  • Dev / staging / prod environments generated from the same code, differing only by variables

Impact

100%

of platform infrastructure captured as code

Reproducible

any environment can be rebuilt from Git

Auditable

every change reviewed, versioned, revertible

Project Gallery

Flagship builds

Filter by discipline — each card is a system running in the real world.

Platform

Kubernetes as a Service

Multi-tenant KaaS platform with self-service onboarding for client environments.

KubernetesMulti-tenancyHelm
Platform

OpenShift Production Estate

Architecture, governance and day-2 operations of production OpenShift clusters.

OpenShiftMCOGovernance
Security

DevSecOps Pipeline Platform

SAST / DAST / IAST security gates embedded in group-standard CI/CD templates.

SASTDASTGitLab CI
Security

Secrets & Compliance Baseline

Centralized secrets management and controls-as-code governance for cloud workloads.

SecretsComplianceWazuh
Observability

Unified Observability Stack

Prometheus, Grafana, ELK, Loki and Instana — one telemetry surface for every environment.

PrometheusGrafanaInstana
Automation

Everything-as-Code Foundation

Terraform + Ansible + GitOps lifecycle for Azure, AWS and hybrid infrastructure.

TerraformAnsibleGitOps
Automation

Group CI/CD Standardization

Reusable pipeline libraries adopted across international entities of the group.

GitHub ActionsJenkinsAzure DevOps
Platform

Hybrid Cloud Landing Zones

Secure Azure and AWS landing zones with HA and disaster-recovery design.

AzureAWSDR

Tech Stack

Depth where it matters

Proficiency is earned in production. Years reflect hands-on operational experience, not tutorials.

Cloud Platforms

  • Microsoft Azure5 yrs
  • AWS5 yrs
  • Hybrid / Private Cloud4 yrs

Containers & Orchestration

  • Kubernetes6 yrs
  • OpenShift4 yrs
  • Docker & Helm6 yrs

IaC & Automation

  • Terraform5 yrs
  • Ansible5 yrs
  • GitOps4 yrs

CI/CD

  • GitLab CI5 yrs
  • GitHub Actions4 yrs
  • Jenkins · Azure DevOps5 yrs

Observability

  • Prometheus & Grafana5 yrs
  • ELK · Loki4 yrs
  • Instana APM3 yrs

Security & Scripting

  • SAST / DAST / IAST4 yrs
  • Linux8 yrs
  • Bash · Python · YAML6 yrs

Tech Radar

How I evaluate the ecosystem right now

Adopt

Default choices — proven in my production environments

  • Kubernetes
  • Terraform
  • GitOps (Argo CD / Flux)
  • Prometheus + Grafana
Trial

Actively running in real projects, watching the edges

  • OpenTelemetry
  • Crossplane
  • Backstage (IDP)
Assess

Prototyping and tracking closely

  • eBPF security tooling
  • AI-assisted operations
  • Platform FinOps tooling
Hold

Deliberately avoiding for new builds

  • Hand-managed servers
  • Click-ops cloud consoles

Global Availability

Based in Madagascar. Built for the world.

UTC+3 — a full working-day overlap with Europe and comfortable coverage of Middle East and African time zones. Fully equipped for remote-first collaboration; open to relocation for the right mission.

AntananarivoParisMauritiusDubaiNew YorkSingapore
Antananarivo · UTC+3

Remote — worldwide

Proven track record of delivery with distributed teams in France and Mauritius.

Hybrid — EMEA

Available for on-site phases: kickoffs, architecture workshops, incident retrospectives.

Relocation — open

Open to international relocation for senior platform and DevSecOps roles.

  • EuropeCore overlap · daily collaboration
  • Africa & Middle EastNative time zone fit
  • North AmericaMorning overlap · async-friendly
  • Asia-PacificAsync collaboration

Work With Me

Four ways to engage

Whether you need a platform built, a team leveled up, or an architecture rescued — there's a format that fits.

Platform & Cloud Consulting

Architecture reviews, Kubernetes/OpenShift design, landing zones, migration strategy. Senior judgment, delivered fast.

  • Cloud architecture & landing zone design
  • Kubernetes / OpenShift platform blueprints
  • DevSecOps maturity assessment & roadmap
Discuss this engagement

Freelance Delivery

Hands-on build engagements: I ship the platform, the pipelines and the runbooks — production-ready, documented, handed over.

  • KaaS / platform builds end to end
  • CI/CD and GitOps industrialization
  • Observability stack deployment
Discuss this engagement

Full-Time Leadership

Principal / Staff platform roles or DevSecOps leadership. I build the platform and the team that runs it.

  • Platform engineering leadership
  • Team mentoring and technical roadmaps
  • Remote or relocation, international scope
Discuss this engagement

Workshops & Speaking

Practical enablement for engineering teams — from Kubernetes fundamentals to production-grade GitOps.

  • Kubernetes & GitOps team workshops
  • DevSecOps enablement sessions
  • Conference and meetup talks
Discuss this engagement

Engineering Philosophy

The manifesto

Platforms exist for one reason: to make the right thing the easy thing. Every design decision I make gets tested against that sentence.

Automation is not a productivity hack — it is the only honest way to operate at scale. Humans review intent; machines execute state. If a change can't be expressed as code, reviewed as a diff and reverted with a commit, it doesn't belong in production.

Security that slows delivery gets bypassed, so I build security that ships: gates in the pipeline, policies in the cluster, secrets in a vault — invisible when you comply, immovable when you don't.

And none of it matters if you can't see it. Observability is the platform's nervous system; cost is a first-class metric alongside latency and error rate. A platform team that ignores FinOps is just moving the outage to the invoice.

Everything as Code

Infrastructure, pipelines, policies, dashboards — versioned, reviewed, reproducible.

Paved Roads, Not Gates

Make the secure, observable path the fastest path. Adoption follows.

Boring Technology Wins

Innovation budget is spent where it differentiates — not on exotic infrastructure.

Feedback Loops Over Heroics

Fast pipelines, honest dashboards, blameless retros. Heroism is a symptom of platform debt.

Credentials

Certifications & Education

Certifications

Certified Kubernetes Administrator (CKA)

Cloud Native Computing Foundation

In progress

Microsoft Azure Solutions Architect

Microsoft

Microsoft Azure Administrator

Microsoft

DevOps & Cloud Engineering Training

Professional certification track

Education

Master's — Systems & Networks

École Nationale d'Informatique

Valedictorian · Highest honors (Très Bien)

Bachelor's — Systems & Networks

École Nationale d'Informatique

Valedictorian · With honors (Bien)

Social Proof

What teams say about working with me

“Hercule doesn't just administer clusters — he designs operating models. He took our fragmented deployment practices and turned them into one standard that every entity of the group actually adopted.”

Engineering Director

Groupe Constellation

“The rare architect who is equally credible in a whiteboard session and at 2 a.m. during an incident. His observability stack changed how we run production.”

Lead SRE

Platform team, IDWEB / AMWEB

“He made security a default instead of a debate. Our pipelines enforce what our policies used to merely describe.”

Security Officer

Group security function

Representative feedback from colleagues and stakeholders — full references available on request.

FAQ

Common questions, direct answers

Yes — I'm open to consulting and freelance engagements immediately, and to full-time senior platform roles for the right mission. The fastest way to check fit is a short intro call.

Insights

Writing & thinking

Field notes from production — platform engineering, DevSecOps and the craft of running systems that don't wake you up.

Platform Engineering Draft in progress

Kubernetes as a Service: what multi-tenancy actually costs

Namespaces are free; isolation is not. Lessons from building a KaaS platform that real clients run on.

8 min read

DevSecOps Draft in progress

Security gates developers don't hate

How to wire SAST, DAST and policy checks into pipelines without becoming the department of no.

6 min read

Observability Draft in progress

Your dashboards are lying to you

Symptom-based alerting, SLOs, and why host metrics are the wrong thing to page on.

7 min read

Contact

Let's build something that lasts

Tell me about your platform, your bottleneck or your role. I reply personally, usually within 24 hours.

Email

herculeihasa@gmail.com

Phone / WhatsApp

+261 34 37 426 63

Location

Antananarivo, Madagascar (UTC+3)

Your message is stored securely and only used to reply to you. Privacy