Build Notes

Project stories and technical notes

Multi-agent systemsAI evaluationProduction AIResearch methodsFounder engineering

Reader-first notesTechnical depth, clear boundariesFiltering: Multi-Agent Systems

Peter McCann Strain working at a laptop on a sunny Glasgow balcony — From the writing desk

All AI Evaluation Computer Vision Founder Engineering LLM Cost Optimization Legal Infrastructure Medical AI Multi-Agent Systems Production AI Recruitment AI Research Methods

Conversico: AI Voice Operations for Dental Practices

Multi-Agent Systems • Production AI

Start here

Start with this

Conversico: AI Voice Operations for Dental Practices

9 min read

An in-production founder case study on an AI voice receptionist for dental practices, built around missed-call recovery, booking support, privacy, and launch-gated operations.

In production · going to market · demo video coming soon

Enter article

Guided entry

Aurum at SThree: Enterprise Agentic AI Engineering

Multi-Agent Systems • AI Evaluation

12 min

Aurum at SThree: Enterprise Agentic AI Engineering

An employer-IP-bounded case study on Aurum at SThree: enterprise agentic AI engineering, reviewable state, fairness diagnostics, red-team testing, production control, and the self-improving-evaluator research that moved scoring quality from a 0.627 baseline to 0.851 without learning the benchmark.

Demo video coming soon

Personal Assistant AI: From Second Brain to Executive Assistant OS

Multi-Agent Systems • Production AI

10 min

Personal Assistant AI: From Second Brain to Executive Assistant OS

A private multi-agent second brain in daily use, with durable memory, proactive support, and quality monitoring behind a privacy-safe engineering trail.

Ongoing case study · demo video coming soon

The Retrieval Bottleneck: A Controlled Comparison of 13 Deep-Research Architectures

AI Evaluation • Research Methods

13 min

The Retrieval Bottleneck: A Controlled Comparison of 13 Deep-Research Architectures

A controlled benchmark of 13 deep-research agent architectures on one shared tool layer — what happens to orchestration claims when you fix the model, the tools, and the judging protocol, and read the statistics honestly.

18 May 2026