Software architect & engineer Independent consultant

Architecture, automation and AI for systems built to last.

01 — Disciplines

Three overlapping practices, one mindset.

— Architecture

Systems that scale with the business, not against it.

Requirements analysis, modular design, integration architecture and event-driven decomposition — done before a line of production code is written.

  • Requirements & process analysis
  • Event-driven architecture
  • Microservices & messaging
  • Integration design
— Automation

Removing the manual work between systems.

API integrations, data synchronisation and asynchronous pipelines that eliminate repetitive operations and keep information flowing reliably.

  • API integrations
  • Data & ETL pipelines
  • System synchronisation
  • Reporting workflows
— Applied AI

LLMs and agents solving real business problems.

Practical integration of language models, retrieval and agentic workflows — only when they measurably reduce cost or unlock new capability.

  • LLM integration
  • RAG systems
  • Agentic workflows
  • Content & data pipelines
02 — Approach

Most failed projects fail before the first commit.

Principles Four working assumptions that shape every engagement.
Most problems in software projects are not technological. They come from unclear requirements, weak communication between business and engineering, and writing code before anyone agreed on what the system is meant to do.
  • 01 — Understand first Process, constraints and future direction are mapped before stack choices enter the conversation.
  • 02 — Bridge two languages Business speaks in goals. Engineering speaks in interfaces. The architect translates — precisely, in both directions.
  • 03 — Design for change Systems that cannot evolve become liabilities. Every decision is weighed against the cost of changing it later.
  • 04 — Technology serves the goal New tools and models earn their place only when they beat the boring alternative on something that matters.
03 — Selected work

Recent projects & systems shipped.

/ 01

Local news aggregation & AI publication platform

Event-driven system pulling, processing and publishing local content via LLM-assisted workflows and social-media automation.

AI · Event-driven
/ 02

Cross-system integration layer with message queues

Asynchronous integration connecting heterogeneous business systems via RabbitMQ, with structured retries, observability and event sourcing.

Integration · Backend
/ 03

Content generation pipeline with language models

End-to-end pipeline for article and social content production using LLMs, with quality gates and human-in-the-loop review.

AI · Automation
04 — Stack

Tools currently in rotation.

Backend
Python
TypeScript
Java
PHP
Data
PostgreSQL
MongoDB
RabbitMQ
Frontend
React
MUI
Vite
AI & Infra
OpenAI API
RAG
Docker
Agent frameworks
05 — Writing

Long-form notes on architecture, security & AI.

Let's design something built to last.