marceau.design

Systems-oriented software builder & consultant

Better systems start with understanding the problem.

I help small businesses untangle awkward processes, disconnected information and gaps between the tools they already use, then design the smallest useful intervention.

Sometimes that means improving a process. Sometimes connecting existing software. And sometimes building something new.

Discuss a problemSee my work

Recognising the problem

When everyday work starts working against you.

The need for change often appears first as friction, uncertainty or workarounds—not as a neat software requirement. Examples of problems I work on include:

  • Repetitive manual work that consumes attention every week.
  • Information spread across spreadsheets, email and separate applications.
  • Systems that work individually but do not work together.
  • Important operational knowledge held mainly in one person's head.
  • Data that is difficult to trust when a decision needs to be made.
  • Reporting that depends on copying, re-entering and preparing information by hand.

A practical approach

Not every business problem needs custom software.

  1. Understand

    Start with the actual problem, the people involved and the way the current system works.

  2. Simplify

    Remove avoidable complexity and improve the process before adding more software.

  3. Connect

    Link useful existing tools or automate the gaps between them where that is enough.

  4. Build

    Create custom software only where it provides meaningful value and earns its place.

Selected work

Building systems one justified step at a time.

Living project

Household Inventory Tracker

A system developed incrementally from a console application with JSON persistence into a PostgreSQL-backed system with console and REST API interfaces. Azure deployment is planned as the next stage.

Its evolution provides practical evidence of backend engineering, data modelling, API development, testing discipline and maintainable system design.

Read the HIT case study

Project evolution

  1. Console
  2. PostgreSQL
  3. FastAPI
  4. Docker
  5. Testing & CI
  6. Azure (planned)

HIT is a living project; Azure deployment is the next planned stage.

Technical foundation

The tools support the work. They do not define it.

  • Python
  • SQL / PostgreSQL
  • FastAPI
  • Docker
  • Azure (learning / planned)
  • REST APIs
  • Testing
  • Git / GitHub
  • CI/CD

About

Business context and hands-on engineering belong together.

My background around enterprise applications, operational processes and data flows informs how I approach systems today.

I pair that perspective with hands-on Python, backend and data engineering: first understanding how work actually happens, then building only what earns its place.

Start a conversation

Tell me what isn’t working.

You do not need a technical specification, and you do not need to know whether software is the answer. Start with the process that feels awkward, repetitive, unclear or harder than it should be.

A useful first description: what are you trying to improve, and where does the current way of working get in the way?

Discuss a problem