borjaglez.com
Available · remote

// node 00 · Ourense, Spain · 8+ years

Borja
González

Software Architect & Senior Full-Stack Developer

I design systems that start cheap and scale horizontally when the product demands it. Eight years building web, mobile and real-time products; now extending into generative AI.

CLIENTSENTRYMESSAGING · CQRSSERVICESDATAcmd · queryeventAngularIonic · Flutter · ExpoExternal systemsAPI Gateway · RESTBus CQRS — RabbitMQ · KafkaSpring BootNode.jsWorkersPostgreSQL · Oracle

fig. 01 — reference architecture · main product● command · query● event

+8
years building software
2
libraries on Maven Central
100 %
test coverage in my libraries
3
open source projects
01 / Profile// context

Fundamentals first, frameworks second.

Software engineer with eight-plus years building web, mobile and real-time communication products. At Imatia Innovation I define the architecture of products and projects from the ground up to production.

Microservices with Spring Boot and Node.js, CQRS, asynchronous messaging and Angular frontends. Every system is load-tested before production and wired into automated CI/CD pipelines.

Spanish · Galician native — English intermediate — Telecommunications Engineering studies, Universidade de Vigo

02 / Experience// timeline

From intern to architect at the same company: Imatia Innovation.

Dec 2021 — present
Imatia Innovation · Vigo

Software Architect · Senior Developer

  • Define the architecture of products and projects from the ground up to production: design, scalability and technology choices.
  • Microservices with Spring Boot and Node.js, CQRS and async messaging with RabbitMQ and Kafka: dozens of services in production.
  • Load testing as a gate before production: every architecture is validated above the expected peak load.
  • CI/CD on Docker Swarm and Kubernetes: from manual deploys to several automated releases per week.
  • Flagship product: high-load backend, Angular, WebRTC video over SFU and iOS/Android apps in daily production use.
Jan 2019 — Nov 2021
Imatia Innovation · Vigo

Software Developer

  • Full-stack development of public-sector products and web apps with Spring Boot, Node.js and Angular, plus hybrid apps with Ionic.
  • Growing involvement in technical design, the basis of the architecture responsibility taken on later.
Sep 2018 — Jan 2019
Imatia Innovation · Vigo

Intern Developer · FEUGA grant

  • Public-sector web application with Java, Spring Framework and JSF (PrimeFaces).
02.1 / Projects and technologies

What I have worked with, beyond the stack.

fig. 02 — WebRTC video over SFU

One stream per peer, N subscribers: how the video call scales.

Each participant publishes a single stream to the SFU, which forwards it to the others without mixing: cost grows linearly and latency stays low. Signaling travels over WebSocket to an independent backend (Node.js or Rust); STUN and TURN handle NAT and restrictive networks.

● media- - signaling

PEERSREAL TIMEBACKENDAngular · webIonic · FlutterExpo · RNSignaling · WSSFU · media routerSTUN · TURNNode.js · Rust
PIPELINEDEPLOYfailcov. 100 %peak loadgit pushbuildtestsload testregistryDocker SwarmKubernetes

fig. 03 — CI/CD pipeline · Docker Swarm and Kubernetes

Every push goes through build, tests, load and deploy with no manual steps.

100 % coverage and load testing are gates in the pipeline, not optional steps: if they fail, the change goes back to the commit. What passes is published as an image and deployed to Docker Swarm or Kubernetes, several times a week.

● change- - failure, back to commit

Cross-platform mobile

Ionic, Flutter and React Native (Expo)

iOS and Android apps for the main product, in daily production use.

Validation under load

Load testing before production

Every architecture is tested above the expected peak load: low initial cost, horizontal scaling.

Public sector

REST and SOAP integrations, maps with Leaflet

Products and web applications for public administrations with Spring Boot, Node.js and Angular.

03 / Open source// artifacts

Public code, 100 % coverage.

Two libraries published on Maven Central and one backend-less web tool. Every build enforces full test coverage; it is the public proof of how I work.

04 / Stack// layers

Tools change; fundamentals do not.

Backend
JavaSpring BootNode.jsMicroservicesCQRSREST · SOAPRabbitMQKafkaWebRTC (SFU)GraalVM
Frontend
AngularTypeScriptReactHTML · CSSViteWebpackEsbuildLeaflet
Mobile
IonicFlutterReact Native (Expo)Android (Kotlin)
Data
PostgreSQLOracleMariaDB · MySQLNoSQLELK
DevOps
Docker SwarmKubernetesCI/CDLoad testingGitAWS
AI
LLMsFine-tuningRAGAgentsLangChainLangGraphVercel AI SDKPythonPyTorch
Other
RustC / C++C#
05 / Generative AI// in progress

Right now: LLMs, fine-tuning and agents.

Extending my architecture profile into generative AI through the IBM specialization: LLM architecture, foundational NLP models, transformers and fine-tuning. Next up: RAG and agents.

IBM · Coursera · professional certificate · starting

RAG pipelines with vector databases, multimodal applications and agents with LangChain, LangGraph, CrewAI and MCP.

Additional training — Complete Rust course (Udemy, 2024) · Kubernetes and Docker on AWS (Udemy, 2023) · Angular: The Complete Guide (Udemy, 2019) · Hibernate and Java Spring (Platzi, 2019)

06 / What I believe// invariants

Five rules I do not negotiate.

01

Fundamentals over frameworks.

If you do not understand the problem, the tool will not save you.

02

Architecture matters.

Systems that start cheap and scale horizontally when the product asks for it.

03

Nothing ships without load testing.

Validating under load is part of the design, not an extra.

04

Testing is not negotiable.

My libraries ship with 100 % coverage enforced on every build.

05

Code you understand on first read.

Clean architecture and maintainability as the criteria for choosing technology.

07 / Split// metrics

Where my time goes.

30 %
Architecture & design
30 %
Backend
20 %
Frontend
10 %
Mobile & WebRTC
10 %
DevOps & CI/CD
08 / Contact// endpoint

Let's talk.

Open to software architecture, senior full-stack and AI roles in Spain, remote preferred. If you have an interesting project or need a second opinion on architecture, scalability or testing, write me.