Technical LeadershipAugust 22, 2026

From Senior Engineer to Solution Architect: My Certification Roadmap 2026–2029

How I structured a 3-year certification path to validate existing experience and build the credentials for Solution Architecture — without collecting unrelated badges or burning out.

Carlos David Duarte

Carlos David Duarte

Senior Software Engineer · Technical Lead

There's a pattern in engineering careers that gets people stuck: accumulating certifications without a thesis. A cloud fundamentals badge here, a security course there, a Kubernetes credential because it was on sale. The result is a profile that looks busy but doesn't tell a coherent story.

I've done this. Most engineers have.

This post is about doing the opposite — designing a certification path that validates where I've already been and builds evidence for where I'm going.

The thesis

The career trajectory I'm building toward is:

Senior Software Engineer → Technical Lead → Application Owner → Solution Architect → Cloud Architect

That's not a wish list. It's the direction I've been moving for years through actual work at Coppel, Zurich Insurance, and several infrastructure-heavy engagements. The certifications don't create this trajectory — they document and validate it for an audience that can't see the code or the architecture decisions.

The principle behind every certification on this roadmap: validate existing experience, don't just accumulate credentials.

What I already have

Thirteen certifications earned before this roadmap. IBM cloud and data fundamentals, UNIR DevOps diplomas (Administración y Automatización, Despliegue Automatizado de Infraestructura), LinkedIn Learning AI courses, a language proficiency certificate.

They're legitimate. They also don't tell a clean story. The next phase does.

The roadmap

Phase 1 — Azure Foundations: AZ-900 (Q4 2026)

The entry point is deliberate. I've been working with Azure for years — Azure DevOps pipelines, CI/CD orchestration, cloud deployments. AZ-900 isn't the challenge; it's the formal starting line.

Why bother with the fundamentals exam when I'm already past fundamentals? Because skipping it and starting at AZ-204 creates a gap in the credential chain. Hiring managers, clients, and review panels look at the sequence, not just the top item. A Solutions Architect without the foundational certification looks like they skipped steps, even if they didn't.

Target: Q4 2026. Study path: Microsoft Learn AZ-900 learning path + John Savill's study cram.

Phase 2 — Azure Developer: AZ-204 (Q1 2027)

This is where the real engineering validation starts. AZ-204 covers Azure App Service, Azure Functions, Azure Storage, Cosmos DB, authentication, and API integration — the exact surface area I work in daily.

The target date is Q1 2027, roughly four months after AZ-900. That's a realistic preparation window while working full-time: Microsoft Learn paths, hands-on labs through Azure App Service and Azure Functions, and the Azure SDK documentation.

Phase 3 — Cloud Architecture: AZ-305 (Q3 2027)

AZ-305 is the primary objective of this entire roadmap. Azure Solutions Architect Expert is the credential that closes the distance between "engineer who works with Azure" and "architect who designs Azure solutions."

The preparation shifts from feature-level knowledge to design-level thinking: identity and access management across multi-tenant architectures, compute topology decisions, storage strategy, networking patterns, disaster recovery design. The Azure Architecture Center and the Well-Architected Framework are the core curriculum here, not just exam prep.

Target: Q3 2027. This one gets three to four months of dedicated preparation.

Phase 4 — Enterprise Architecture: TOGAF (Q4 2027 – Q1 2028)

TOGAF is a different kind of credential. Where Azure certifications are vendor-specific and technical, TOGAF (Foundation and Practitioner) is vendor-neutral and governance-focused. It covers the Architecture Development Method, business/IT alignment, architecture governance, and stakeholder communication frameworks.

Why include it? Because the path toward Enterprise Architect requires a shared language with non-technical stakeholders. TOGAF is that language — it's how you discuss architecture decisions in a boardroom as well as in a design review.

Foundation: Q4 2027. Practitioner: Q1 2028.

Phase 5 — DevOps Leadership: AZ-400 (Q2 2028)

I've led DevOps transformations, built CI/CD pipelines, and owned delivery infrastructure. AZ-400 validates that experience formally and deepens the Azure DevOps + GitHub Actions surface area.

The timing — mid-2028, after cloud architecture and TOGAF — is intentional. DevOps at the architect level is about delivery system design, not just pipeline configuration. Having the architecture credentials first means the DevOps certification gets interpreted in that context.

Phase 6 — Engineering Excellence: OCP Java SE (Q3 2028)

Oracle Certified Professional Java SE Developer. This one is about anchoring the engineering foundation that everything else is built on.

Java has been my primary language for the bulk of my career — Spring Boot microservices, enterprise application modernization, backend architecture. OCP validates that depth at an internationally recognized level. It's also a credential with a clear signal: people who pursue OCP are serious engineers, not framework-level practitioners.

Phase 7 — Cloud Native: KCNA (Q4 2028)

Kubernetes and Cloud Native Associate from the Linux Foundation. Not because I'm moving toward platform engineering, but because cloud-native infrastructure is now part of every serious architecture conversation.

KCNA covers Kubernetes fundamentals, cloud-native observability, GitOps, and the broader CNCF ecosystem. For a Solution Architect working with enterprise teams, understanding what the infrastructure team is building and why their decisions affect your design is essential.

Phase 8 — AI for Architects: AI-900 (Q1 2029)

The final certification on this roadmap is a deliberate positioning for what comes after 2029. AI-900 covers Azure AI services, machine learning concepts, computer vision, NLP, and AI fundamentals from an architecture and governance perspective — not a data scientist perspective.

The goal isn't to become an AI engineer. It's to make architecture decisions involving AI services with the same confidence I bring to decisions about compute, storage, and networking.

The math on time

Nine certifications over roughly 30 months while working full-time. That's one certification every 3–4 months, with natural spacing between phases that increases as certifications get harder.

The sequencing matters as much as the schedule. Each phase builds on the previous one:

AZ-900 → AZ-204 → AZ-305 → TOGAF → AZ-400 → OCP Java → KCNA → AI-900
  ↓          ↓         ↓        ↓        ↓         ↓         ↓       ↓
Vocab    Dev skills  Arch   Governance DevOps  Core Eng  Infra   AI context

Skipping or reordering this sequence loses the compounding effect.

What runs in parallel

Certifications are evidence. The actual capability development happens through continuous learning alongside the exam prep:

  • Technical writing — at least one architecture-focused article per month on this blog. Writing forces clarity that exam prep doesn't.
  • Architecture diagrams and case studies — applied practice on real or realistic problems, not just exam scenarios.
  • GitHub projects — working code that demonstrates the concepts the certifications cover.
  • LinkedIn — sharing progress publicly creates accountability and surfaces opportunities.

The blog is part of the strategy, not a side project. Each certification phase generates content: what the exam actually covers, how it connects to real work, what surprised me.

What I'm not doing

Not chasing every relevant certification. Google Cloud, AWS, Terraform, CKA, CISSP — all have merit. None are on this roadmap because they dilute the Azure + Java + Enterprise Architecture narrative that matches my actual trajectory.

Not cramming for resale. Certification mills that prep you to pass exams you don't understand produce credentials that don't survive the interview. Every exam on this list corresponds to technology I work with or will work with.

Not treating this as the primary work. The certifications are the proof of record. The actual work — architecture decisions, engineering leadership, delivery ownership — is the career. Exams document it; they don't replace it.

Where this leads

By Q1 2029, the credential portfolio tells a clear story:

Azure cloud architecture (AZ-900 → AZ-204 → AZ-305), enterprise governance (TOGAF), DevOps leadership (AZ-400), core engineering depth (OCP Java), cloud-native fluency (KCNA), and AI architecture context (AI-900).

That's not a random collection. It's evidence for a specific professional position: a Solution Architect with enterprise engineering foundations and cloud-native depth.

The work starts in Q4 2026 with AZ-900. The story is already being written.

More articles

Writing on Java, Azure, DevOps, and engineering leadership.

← Back to Blog