Research programme

Six questions we keep working on

Each thread is a durable area of inquiry. Articles add evidence, calculations, and implementation lessons without pretending that one post settles the question.

THREAD 01

Agent Architectures

How autonomous systems plan, preserve state, recover from failure, and remain inspectable over long-running work.

Explore the thread →

THREAD 02

Language Models

Inference economics, retrieval, model routing, and the boundary between frontier and specialised models.

Explore the thread →

THREAD 03

Agents in Production

What separates a compelling agent demo from a secure, observable, resilient production system.

Explore the thread →

THREAD 04

Voice Systems

Latency budgets, turn-taking, interruption handling, and streaming inference for natural voice interaction.

Explore the thread →

THREAD 05

AI-Native Topology

How teams, controls, and economics change when autonomous systems become part of the operating model.

Explore the thread →

THREAD 06

Externalities & Infrastructure

Power, cooling, silicon, water, and the physical constraints behind training and inference at scale.

Explore the thread →