Aden Mann: Operator-grade
AI.

I build applied AI systems for work where being wrong is expensive. I work independently now; before that, four years at Immutable, a US$2.5B web3 company, building its AI and automation capability out of finance operations. Eleven years flying Army helicopters and an MBA came first.

The main theme for me is operations at the hard end: systems that break at bad moments, decisions with real stakes, and the gap between what a tool is supposed to do and what it actually does under pressure. Army aviation went digital partway through my career. Watching that transition up close, including unpacking where it went wrong as an aviation safety investigator, is where I developed a view on how human-machine work actually transfers, and how it fails.

See the builds ↓

01 Builds

AutoEvaluation

An optimisation loop that hill-climbs LLM instructions against a scoring function. Each round makes one change, keeps it if the score goes up and reverts it if it doesn't. Unlike DSPy, TextGrad and MIPRO, it doesn't need a Python pipeline.

github.com/AdenCJM/AutoEvaluation ↗

AutoEvaluation's experiment log: eight runs, each marked keep or discard next to the change it tried — AutoEvaluation experiment log (demo data)

Parallel Research

Sends one research question to Claude, GPT, Gemini and Perplexity at the same time, then compares the answers to show where they agree and where one model is out on its own.

github.com/AdenCJM/parallel-research ↗

AI Fluency Framework

A five-level, six-function matrix for measuring AI capability across a company. I built it while rolling out AI at a scale-up, after seat count stopped being a useful metric: a licence shows who's paying for the tool and nothing about who can use it.

Read the framework →

+ 10 more builds →

02 Live agent

Ask AI Aden

An agent that answers the way I would, running live on this page.

Live

“How do you know a prompt change actually made things better?”

I score it. Rule-based checks cover anything you can count, like format or banned phrases, and an LLM judge covers the parts that need judgement. If the combined score drops, the change gets reverted automatically.

03 Writing

What Aviation's Automation Transition Tells Us About AI at WorkFeatured2026-04-05 Why Claude Is Different2026-03-10 Workflow Archaeology2025-11-15 AI Barnacles2025-02-28 All writing →

Spoken at / featured in

Speaker on applied AI and AI strategy, New York, Sydney and Melbourne

If you're working on something in this space, I'd like to hear how you're approaching it. I'm open to advisory and fractional work, and the right full-time role.

aden@adenmann.com →