The Future is Agentic - Agentic Literacy:

​What does “Agentic” mean?

“Agentic” or “Agent centric” refers to our world of work moving from a very labour intensive world both physical and cognitive two of the world were all labour both big and small, whether physical or cognitive is being outsourced by cognitive agents or physical robots with a cognitive agentic architecture. So agentic refers to a world where we use agents and agent workflows two complete tasks, goals and work as they can work 24/7 are always highly motivated, don't need sick days and will make your life so much more fun and easier.

It is not feasible for us to do labour going forward as AI would just be able to do it faster for longer and better than we ever can and therefore, us acquiring a new skill set which we call “Agentic Management” or “Agent Management” is key to stay not only competitive but also free ourselves up for high-level thinking, to find better ways of doing things and gaining more overall joy and meaning in our careers and our private lifes. 

AGI (Artificial General Intelligence) will be built from a multitude of agents and agent architecture therefore cementing the agentic future and therefore making agentic literacy the most important skill we can all familiarise ourselves with.

All large companies that are building generative AI platforms (Open AI, Microsoft, Google, Meta, Amazon, Apple & NIVIDIA), and the open source community and the rest of the world have understood and are now focusing on building both “Task Driven Agent” (Microsoft's Copilot & GitHub Copilot Workspace, Open AI GPT, Google's Astra Project & agent builder, Amazon AI Bedrock agents, ) & “Goal Driven Agents” (Microsoft's Autogen, Microsoft's J.A.R.V.I.S, Autogroq, NIVIDIA foundation agents, the open source community with AutoGPT, BabyAGI, SuperAGI, ChatDev, CrewAI, Langchain Agents, LangGraph Agents).

So the future will be agentic because there is humongous value to gain as everyone can build their own workforce making you incredibly more productive than you can ever imagine, so in the rabbit hole we go.

Why Agents?

If you are prompting ChatGPT depending on your prompt engineering architecture and skill and or technique you will get a terrible (hallucinating), good or great output, but if you are using an “Agent Prompting Technique” or one of the many “Task or Goal Driven Agent Frameworks” your output will be extremely good because using expert personas or combining different expert personas Will not only give the large or small language model or the large action model, more context, insights and data then you can ever get from a simple or complex prompt.

So by using agents we can give the large language model more opportunities to answer our questions correctly and with way better output because we are giving LLM just more context, tools, skills and a mixture of different expertise which we normally give with an even more complex prompt.

​

The power of AI comes from combining different experts of AI's. 

​​

​

Agentic Systems Capability Increase:
As of April 2025 AI Agents are currently doubling their capabilities every three month and the length of task they can do every 7 month:

​

​
​

https://www.lesswrong.com/posts/deesrjitvXM4xYGZd/metr-measuring-ai-ability-to-complete-long-tasks 

​

​

5 Stages to AGI:

If some of the most innovative and largest companies are working towards AGI (Artificial General Intelligence) & ASI (Artificial Super Intelligence) and they having made progress on those below road map stages and we have clear signals from the leaders that they will have the words best coders as an LLM by the end of 2025 it becomes very clear what we all need to focus on.

​

​

​

​

​

​

​

​

​

​

​

​

​

What to focus on?

Since it is clear that the future is going to be agentic and that:

1. Agents can use Computers (Manus AI, Operator, ACE Agent, Claude Computer Use etc.)

2. Agents becoming extremely reliable by the end of 2025 as advanced reasoning & with the OpenAI building the number one coder as an LLM by the end of 2025

3. This will make multi agent swarm orchestration working and feasible 

4. Agents are going to build & improve agents (Self-Improving)

5. Research agents are going to come up with new novel & new ideas to solve problems which we have just seen with the solving of Potts Model (https://arxiv.org/pdf/2503.23758)

6. AI/Agents to can do the work of organisations

​

So by now it should be crystal clear what is going to happen and what to focus on. 

While it is important as a fundamental core skill to stay:

• A) Up to date with the latest in AI and how your work, skills and personal use, training and development will change as well as knowing

• B) Model, Tool, Prompt Engineering Literacy and becoming

• C) AI savvy and

• D) knowing what to learn next 

​

It is even more important to:

To know how to conceptualise what your armies & your orchestration management will 5, 10, 100, 1000x your output potential and free you up time for higher level thinking and to be more human (connect with empathy with others & uplift to help steer this future.

​​​​​​

Future Skills Sets:

• Agentic Literacy

• Agent Management

• Agent Memory Management

• Agent Skill & Tool Management

• Agent Testing & Monitoring Management

• Agent Model Usage & Workflow Management

• Agent Workforce Architecture & Orchestration Management

• Multi Agent System building & Conceptualisation of 5-1000x your output

• Learning & Self-Improvement Agent Management using new orchestration Frameworks 

• Grid & Resilience towards disruptions that this transformational time will bring as a Skill

• Constant Up-skilling on technologies that your organisation has not adopted yet as a Skill 

• Being able to imagine what your next steps are and how you orchestrate 10-100's of agent in a coherent swarm to do massive amount of work

​

​

​​​​

​