Generative AI makes things for you. Agentic AI does things for you.
This distinction, between making and doing, is the next step on the exponential curve of capability.
An AI Agent can initiate and execute tasks independently, finding the best way to transition from an initial state to a goal state by using any tools it has available, as often as it is capable of doing.
In his excellent primer on AI Agents, Agentic Design Patterns, Antonio Gullí defines the agentic loop in 5 stages:
- Get the mission.
- Scan the scene.
- Think it through.
- Take action.
- Learn and get better.
When you understand the mechanics of the agentic loop (observe, reason, and act) you can stop prompting for responses, and begin composing loops instead.
This is something I help intermediate-to-advanced AI operators do in my 300-level workshops. I am hosting an 8-week sprint to ship agentic workflows, this month in Christchurch in the AI Agent Accelerator.
- 1 October: Ai301 – Context Engineering
- 8 October: Ai302 – Agent Management
- 15 October: Ai303 – Vibe Coding Applications
- 22 October: Ai304 – Agent Harnesses
- 29 October: Ai305 – S.C.O.P.E. Agent Tasks
- 5 November: Ai306 – Automate Your Workflows
- 12 November: Ai307 – Generate System Prompts
- 19 November: Ai308 – Evaluate Agent Performance
These 8 workshops will be held on Thursday mornings, 10am-11:30, at the new EPIC Exchange.
During Christchurch AI Meetup, the Head of EPIC Innovation (our incredible venue in Christchurch) announced the EPIC Exchange, a collaborative coworking space focused on cool people doing cool things with AI.
I am going to be one of the Founding Members, and for my first 12 colleagues, they will get free access to the AI Agent Accelerator, an 8 week workshop sprint.
🗂️ “An AI Agent runs tools in a loop to achieve a goal.” – Simon Willison
If you don’t want to join the EPIC Exchange, you can still enroll in the upcoming 8-workshops series for $350 NZD.
If you’re not in Christchurch, but you’d like to join a webinar sprint on these 300-level topics, you can indicate your interest by clicking here, and I will follow-up to find times and dates that work for interested attendees.
🎬 Interview with Peter Cook
[14:29]
👓 What I’m Reading
- We Must Pace the Frontier by Dario Amodei – “The current models are an almost endless gold mine of insight into both how to build AI well and what can sometimes go wrong with it if it isn’t built well. I believe that if slowing down bought us even an extra year or two before models reach critical levels of capability, and we used that time to advance alignment, we could greatly reduce the risk that something goes seriously wrong.”
- open the frontier by Jack Dorsey – “forcing someone to withhold a general-purpose model for safety reasons is a last resort. i would support it only with independently reviewable evidence that release materially increases a risk of catastrophic harm that narrower measures cannot adequately address.”
- P(doom) calculation by Hunter Ash – “Doomer arguments are an illusion of rigor. You can string together a cause-and-effect story and attach vibed numbers and reach any conclusion you want. I’m not saying people don’t really believe their stories, but they are still just stories.”
- What if AGI is already here, and it’s made of… children? by Pascio – “In biology, there’s a specific name for ‘slight reinterpretation across generations of self-replicating entities.’ We call that mutation. And mutation, combined with selection pressure, is… well, the very engine that produced every single intelligent thing on this planet today.”
Leave A Comment