Gareth Austen AI systems · product engineering

Notes from the workbench

I build AI systems that have to work in the real world.

I’m Gareth Austen, a software and machine-learning engineer in Dublin. This is where I write about optimising Agents, developing robust quality gates, designing good evals, building Tell Me Ted, working with coding agents, and turning prototypes into dependable products.

Latest writing

All writing

  1. Building a Physical Product When Your Background Is Software

    Schematics, components, batteries, suppliers, prototypes, certification and manufacturing: lessons from taking a software-led idea into the physical world.

    Building Tell Me Ted· Embedded Systems· Manufacturing

  2. Why Our First AI Architecture Was Too Complicated

    Why an architecture built around separate AI capabilities became too fragmented for the continuous reading experience we wanted.

    Building Tell Me Ted· Artificial Intelligence· Voice AI

  3. How We Turn Product Ideas Into Codex Tasks

    A repository-aware workflow for converting product ideas into Codex tasks with invariants, tests, review prompts, device validation and rollback instructions.

    Building Tell Me Ted· Codex· Embedded Systems

Series in progress

Building Tell Me Ted

Twelve field notes on building a physical AI product: the architecture, the hardware, the judgement calls and the mistakes.

  1. 01 We’re Building a Teddy Bear That Can Read With a Child
  2. 02 Building a Startup With ChatGPT as a Daily Collaborator
  3. 03 Codex Can Write the Code. You Still Own the System.
  4. 04 What’s Inside an AI Teddy Bear?
  5. 05 From Chained AI Calls to a Realtime Voice Experience
View the complete series