Chris Parsons

The new normal. https://www.chrismdp.com/comics/black-october/

Might have reached my cognitive limit in a working session last week.

I'm building a language for describing who does what in a workflow and now I can understand workflows better, Claude can work so much harder on them. The main limitation for my productivity is me, then the tools I have. As we build better tools, the sky's the limit (or my brain explodes).

https://www.chrismdp.com/awml/

chrismdp.comYou Must Draw Your AI Workflow to Control ItI have been building my own workflows, working in tandem with AI, for over a year now. Like many others, I started by giving Claude Code access to my existin...

Might have reached my cognitive limit in a working session last week.

One Claude implementing AWML on my own system, mapping out two workflows at a time, and building a third into a new secure environment. Another Claude is working through the spec, building AWML workflows to help build... AWML - fastest ever software factory I've put together.

The third is building out the AWML registry web app as another factory, trying frantically to keep up with the spec changes.

I don't think an AI rollout starts with permission to use the tool. It starts with changing the work around it.

If every handoff and approval remains the same, faster drafts arrive at the same manager's desk. That manager's day fills up while the team is told it has been transformed. Pick one piece of work. Decide which decisions still need a person and which handoffs can go.

https://www.chrismdp.com/ai-must-be-line-managed/

chrismdp.comAI Must Be Line ManagedI used to just code. Then I managed teams. Now I find myself needing management skills again, even when I am not managing people.The first time I tried deleg...

I'm sorely tempted to make a game again.

I used to build games professionally and as a hobby. Last year I tried using AI to make isometric assets for a gardening game. The pictures looked lovely. The tiles didn't line up, so I couldn't use them.

That's the maddening thing about game development: an image can look right and still be wrong for the job. Scale, alignment and consistency matter more than a convincing screenshot. I'm tempted anyway.

I've been reading about the new Claude Duet - it's another place you might run your AI work. It's a good test for the idea behind AWML: describing the work separately from the platform.

Duet may be one way to do parts of that work. Its research preview can check connected apps, prepare a reply and ask before sending it. But the job belongs to your business, not to Duet.

That's why I made AWML: to draw the steps, the people, the AI and the approval boundaries before choosing where they run. Duet goes in the platform view. It isn't the workflow. https://www.chrismdp.com/awml/

I've been reading about the new Claude Duet - it's another place you might run your AI work. It's a good test for the idea behind AWML: describing the work separately from the platform.

Say you want an AI to chase unpaid invoices. What starts the check? Which record holds the amount due? Who approves the message before it goes to the customer? What happens if the job stalls?

I don't trust AI workflows I can't draw, yet I had no way to draw one! I invented AWML to get around that.

Through this process I spotted a prompt injection vulnerability in my sales funnel agent, and discovered I couldn't tell where my data was coming from in all cases. Amazingly helpful. Give AWML a try on your workflows, and give me feedback! https://www.chrismdp.com/awml/

I’ve open sourced AWML, a modelling language for agentic workflows.

Modelling workflows independently of eg Cowork, Codex, Managed Agents etc gives you auditing, automatic optimisation, and intentional security boundaries. AWML shows who does each step, where information crosses permission boundaries, and which platform runs it.

github.comGitHub - chrismdp/awml: AWML, the Agentic Workflow Modelling Language: describe an agentic workflow in a small YAML file and draw it as a diagram of who may see each part and where each part runsAWML, the Agentic Workflow Modelling Language: describe an agentic workflow in a small YAML file and draw it as a diagram of who may see each part and where each part runs - chrismdp/awml

This is early stage so very keen to get feedback - comment or open a GitHub issue. GitHub: https://github.com/chrismdp/awml Blog: https://chrismdp.com/awml

A successful AI demo tells you very little about what's going on under the hood, and who's checking it. https://www.chrismdp.com/seven-ai-predictions-from-1882/

I use AI to help with writing and thinking, but I must stand by each word I put out.

chrismdp.com(I Think Whilst) Writing With AIFor thousands of years we have treated writing as proof of thinking. You could not produce a fluent, well-argued piece without doing some thinking along the ...

My new article explains where I draw that line. https://www.chrismdp.com/writing-with-ai/

Instead of selling books, will we be selling skills in the future, with the book coming free with the skill? https://www.chrismdp.com/slice-your-job-into-skills/

AI is good enough, but rollouts stall, and it's normally due to people problems.

You can connect the two: share what the team learns, without asking everyone to abandon their existing work. It sounds obvious, but you have to make sure it happens!

oredev.orgØredev 2026Øredev Developer Conference 4-6 Nov 2026, Malmö, Sweden. Øredev is the Premier developer

This is one of seven people problems in my talk at Øredev in Malmö this November. If you're going, come and tell me how your team is handling it! https://oredev.org/program/25377809-7200-45af-9117-143ee68e15f2

AI is good enough, but rollouts stall, and it's normally due to people problems.

An example: put your best people in a small AI team and they'll move fast. The rest still have a product to maintain. They watch the new work race ahead while their own chance to learn gets postponed.

I'm trying to learn what small teams hoped AI would make easier, and where the impact still isn't big enough.

If your small business is paying for AI, what did you hope would change, and what actually changed? Reply here or message me on Bluesky.

I wondered what it might be like to create video summaries from blog posts, so I tried it. Results below.

David’s point about reviewing early draft outputs is a good one!

chrismdp.comSeven AI Predictions From 1882In the late nineteenth century, J. P. Morgan, already one of the most powerful bankers in New York, wanted electric light before anyone else had it. So Thoma...

I’m wondering whether to build this into my blogging process now. Feedback welcome: am I just contributing to the avalanche of slop out there, or is this a good way to reuse content? https://www.chrismdp.com/seven-ai-predictions-from-1882/

I wondered what it might be like to create video summaries from blog posts, so I tried it. Results below.

It took five minutes of my time to get this 80% good enough, but another hour and a half of my effort (and much longer for Claude) to get it to 99%. I had to rewrite the script after seeing the first version, which cost a lot of wasted tokens.

Hi - I'm Chris 👋 thought I'd post a bit about me and how I got into what I do now. I've realised my whole career can be summarised in seven year blocks:

I'm building products at the same time: 27agents helps orgs get started with ready made AI agents. My mission is to help ensure we don't screw up AI as a society. I want to demystify AI, help people get comfortable with and take advantage of the big changes it will bring.

Introduce yourself - would love to hear from you! Photo at the top was taken on my silver wedding anniversary trip to Dubrovnik in August 💍

Hi - I'm Chris 👋 thought I'd post a bit about me and how I got into what I do now. I've realised my whole career can be summarised in seven year blocks:

I then spent seven years in founding / exec roles at startups. I scaled Gower Street a data science film startup as CTO and then co-founded Cherrypick with Tom during Covid - I'm still an advisor there.

I went full time on AI consulting last year, and this will consume at least the next seven years. I help CEOs and execs like me get value from AI in their organisations - advisory, workshops, briefings, troubleshooting and strategy. It's lots of fun.

Hi - I'm Chris 👋 thought I'd post a bit about me and how I got into what I do now. I've realised my whole career can be summarised in seven year blocks:

My wife Ellie and I moved out of London in 2003 to start a family, and I spent the next seven years founding a client services agency. We managed to bootstrap all the way to $1MM in ARR before the financial crisis hit hard, and we had to close.

The next seven years were itinerant. I contracted at Government Digital Service, ran courses on great software development, and build a video game on Steam in my spare time. Working on lots of things at once with a primary school age family kept me busy!

Hi - I'm Chris 👋 thought I'd post a bit about me and how I got into what I do now. I've realised my whole career can be summarised in seven year blocks:

0-7: uni + games 7-14: agency founder 14-21: itinerant contractor + trainer 21-28: exec + founder 28+: AI consulting My first seven years were at university and in video games in the early 2000s as an AI programmer, when AI meant "follow these rules" - how times have changed!

Are you keeping a knowledge base (aka a "shared brain") yet, for you and your team?

I then decide what the argument is and where I want to take the article. I wrote the whole process down last week.

How I write with AI: https://www.chrismdp.com/writing-with-ai/ What to put in your knowledge base: https://www.chrismdp.com/what-to-put-in-your-ai-knowledge-base/

Are you keeping a knowledge base (aka a "shared brain") yet, for you and your team?

They are extremely useful! My AI can find things I said in a meeting months ago and cross reference with the latest stream of emails.