Engineering Journey

Automation, reuse, and platform thinking at scale

AB INBEV // ACTIVE

At AB InBev, the focus became larger systems with longer operating horizons: automation that can run for years, a frontend design system plus reusable engineering foundations for multiple teams, Nx-based monorepo and microfrontend patterns for scale and release management, and now an AI-centered solutions platform that combines React frontends, Python services, and agent-based workflows to improve both product capability and developer productivity.

platform/current-focus.ts

MVP Window

3 months

Automation Runtime

2 years

lightweight JSON maintenance

Current Focus

AI solutions

React + Python + agents

01 automation + ai

Signal intake

scraper jobs, product requests, team prompts

02 agents

Agent orchestration

CrewAI flows coordinate tasks and reasoning

03 backend

Domain services

Python and FastAPI services apply business rules

04 quality

Evaluation

validation, checks, and typed contracts gate outputs

05 frontend

Handoff

React applications expose the final operator workflow

platform/current-focus.ts

const currentPlatform = {
  automation: createPipeline({
    runtime: ['NestJS', 'Puppeteer', 'BullMQ'],
    storage: ['Redis', 'SQLite'],
    maintenance: 'JSON-driven rules',
  }),
  aiSolutions: {
    frontend: ['React', 'React Router v7', 'Zustand', 'React Hook Form', 'Zod'],
    backend: ['FastAPI', 'SQLAlchemy', 'Alembic', 'PostgreSQL'],
    agents: ['CrewAI', 'Python domain services'],
  },
};

Platform composition

scraper plus ai solution stack

Automation

Long-running data pipelines

NestJS Puppeteer BullMQ Redis SQLite

Frontend

Operator-facing AI surfaces

React React Router v7 Zustand React Hook Form Zod

Backend

Domain and persistence layer

Python FastAPI SQLAlchemy Alembic PostgreSQL

Agents

Current AI workflow focus

CrewAI Domain services Evaluations Nx utilities ESLint

Foundations

Shared delivery standards across teams

Design system Shared templates Utilities Validation layers OpenAPI middleware Nx monorepo Microfrontends

NestJS Puppeteer BullMQ Redis SQLite React React Router v7 Zustand Zod React Hook Form ESLint PostgreSQL Python SQLAlchemy FastAPI Alembic CrewAI Nx

From self-taught builder to technical lead

Creze, Avanttia, Kavak, and Solidus Capital shaped the transition from hands-on developer into architecture and leadership. The common thread was speed with structure: learning unfamiliar stacks quickly, translating business goals into technical execution, and creating clearer systems so teams could move with less friction.

React Vue Angular Node.js AWS Cognito Microservices

workspace/blueprint.txt

workspace/
  apps/
    shell
    pricing
    analytics
  libs/
    ui
    api-contracts
    validation

// reusable foundations for teams and faster delivery

Growth indicators

Leadership and delivery focus

Delivery speed High

DX focus Strong

Support, Linux, automation, and the path into software

The earliest professional stage was not glamorous, but it was formative. Support work built deep systems intuition across Linux, telephony, networking, and troubleshooting. Repetitive operational problems gradually turned into scripts, dashboards, and automation, which made the move into software engineering feel less like a jump and more like a natural extension of solving problems.

foundations/automation.py

for ticket in incidents:
    diagnose(ticket)
    automate_if_repeatable(ticket)

deploy('python-scripts')
learn('django', 'linux', 'networks')
build('from support into software')

Linux Asterisk Python Django CLI tools

2022