The 1, 2, 3 Approach for All Gen.AI Work.pptx

Generative AI is sweeping the world and advancing human capability in almost everything professions and lifestyle. In this presentation I share the underlying method for getting the most out of ANY GENERATIVE AI TOOL. A simple to follow technique, the 1-2-3 Gen. AI Approach is for any skill level to get the most out generative tools.

The prompt has become a crafting of your intent to better align with the thinking of a machine. However, with LLMs that thinking is now more or less the logic we would use with a human colleague.

With computational power now closer to the spoken language, we can converse with things rather than iterate through test problems. But the steps are basically the same.

Leading with AI is not intended to be perfect from whatever it generates. It will produce something, but often it will lack something, have gaps it’s logic, or even present something askew that requires correction.

If you want to stay ahead of the next generation of work, thenthe 1-2-3 Gen. AI Approach will be a cornerstone in your professional and domestic development.

Check it out and drop a comment on how this helped you. Thanks, and enjoy!

1. The 1-2-3 Gen. AI Approach A simple system for working with Generative AI By Jonathan Essary Dialexa AI Summer Camp 2024

2. About Me

3. AI Learning Lessons Using Gen.AI is as easy as 1,2,3! Start w/ prompt structure Process or Flow steps Post Editing Guidance. Mad-lib your starting prompt and make their own. Think through the step-by-step flow of requests to generate your desired outcomes. Consider the necessary post edit things for the the outputs to align with the a quality result.

4. AI Learning Lessons 1,2,3 Example Start with Prompt Structure Mab-Lib the Starting Prompt The prompt has become a crafting of your intent to better align with the thinking of a machine. However, with LLMs that thinking is now more or less the logic we would use with a human colleague. Machine logic like loops, explicit references, and binary actions like John Maeda references in his book How To Speak Machine. However, the way we can access those computational powers is not very similar to our spoken language. Process or Flow Steps A Guided Conversation Post-Edit Guidelines Apply your expertise for the final result With computational power now closer to the spoken language , we can converse with things rather than iterate through test problems. But the steps are basically the same. Just like working through an idea with a colleague, you can guide the output, intent, and discourse toward a better understanding of the problem, better formatting of the output, and build on the knowledge gathered and aggregated over the conversation. Leading with AI is not intended to be perfect from whatever it generates. It will produce something, but often it will lack something, have gaps it’s logic, or even present something askew that requires correction. Generated content will still need your expertise to correct, validate, and frame the outputs correctly. The post-edit of every gen. AI output is what experts will now spend more of their time doing. 1 2 3

5. W hen applied correctly, this will allow experts to find better solutions in the same amount of time as before, by exploring more ideas , synthesizing more data, and failing faster. AI Learning Lessons

6. AI Learning Lessons Prompt Structure 1. Mab-Lib the Starting Prompt Involved with initial FinancialForce migration and current implementation. 2. A Guided Conversation Involved with initial FinancialForce migration and current implementation. 3. Post-Edit Guidelines Involved with initial FinancialForce migration and current implementation. Rich Theil, Product Forge

7. 1 2 3 4 5 6 AI Learning Lessons A Guided Conversation Rich Theil, Product Forge Introduction and Context Setting: Greet the other person and establish a positive tone. Clearly state the purpose of the conversation. Provide any necessary background information or context. Problem Identification: Ask open-ended questions to understand the problem from the other person's perspective. Listen actively and empathetically. Summarize and confirm your understanding of the problem. Exploration of Solutions: Brainstorm possible solutions together. Discuss the pros and cons of each potential solution. Encourage the other person to share their ideas and suggestions. Decision Making: Evaluate the feasibility and impact of each solution. Reach a consensus on the best course of action. Ensure that both parties agree on the chosen solution. Action Planning: Define clear and specific steps to implement the solution. Assign responsibilities and set deadlines. Discuss any potential obstacles and how to overcome them. Closure: Summarize the key points and agreed actions. Confirm mutual understanding and commitment. End the conversation on a positive note.

8. AI Learning Lessons Post Edits Discover / Design Engineering Product Validation of intent with stakeholder or users Brand color/theme alignment UI functionality or steps to be included Code arrangement and shape to align with current product standards Testing and refining of code Integration testing with current codebase Feature AC and stakeholder prioritization alignment Tone and persona correction Grammar and intent validation

9. AI Learning Lessons

10. AI Learning Lessons Thank you!