Generative AI continues to be a hot topic discussion. People are using generative AI to help with many things, but just like with any technology it's important to understand it's a tool. In this episode of the AI Today podcast hosts Kathleen Walch and Ron Schmelzer discuss how to actually put Generative AI into the apps and products that you use including how to use it in production.



Show Notes:




CPMAI Training and Certification



AI Today Podcast: Generative AI Series: Generative AI & Large Language Models (LLMs) – How Do They Work?



AI Today Podcast: Generative AI Series: Foundation Models, Fine-Tuning, and Domain-Specific LLMs



AI Today Podcast: Generative AI Series: The Drawbacks and Challenges of Generative AI



FREE Intro to CPMAI mini course



Trustworthy AI Group Workshop




This episode is part of our Generative AI series so make sure to subscribe to AI Today to get notified of all upcoming episodes in this series.





AI Today Podcast: Generative AI Series: Implementing Generative AI in production

Technology

Tech News

Podcasting

Gadgets

Software How-To

Experimenting, testing, and refining your prompts are essential. The journey to crafting the perfect prompt often involves trying various strategies to discover what works best for your specific needs. A best practice is to constantly experiment, practice, and try new things using an approach called “hack and track”. This is where you use a spreadsheet or other method to track what prompts work well as you experiment. In this episode of AI Today hosts Kathleen Walch and Ron Schmelzer discuss hack and track in detail.



Keeping track of prompts



It's rare to get the desired response on your first attempt. An iterative process of testing different prompts, analyzing the responses, and then tweaking your approach allows you to gradually hone your technique. Another challenge is that LLMs are constantly evolving. The performance of LLMs is very much domain and task dependent, and the performance will change over time. A current prompting best practice is to use a spreadsheet or other method to track what prompts work well as you experiment.



How to set up your Hack and Track Spreadsheet



Keeping track of prompts that work best for you in which situations, including which LLMs are providing the best results for you at that time, can be incredibly helpful for your colleagues as well. There are many LLMs, and at any one particular time, one LLM may perform better than another in a given situation. Without keeping track of prompts you've written and tested, it's hard to have others try to use these prompts themselves.



When creating a spreadsheet to keep track of prompts, the details matter. Every spreadsheet may be set up a little differently but you'll want to include some essentials. Criteria you can use when setting up your hack and track sheet include: Name of the task or query, Prompt Pattern(s) used, LLM used, date last used for this prompt, Prompt chaining approach used if any, and maybe the person or group that created the prompt.



Kathleen and Ron discuss their own experiences with hack and track in this episode and how learning from others is so critical. By seeing how others are writing prompts it helps you get creative an think of ways to use LLMs you may never have thought of. It also lets you see how others at your organization are writing prompts and the results they are having.



Show Notes:




Free Intro to CPMAI course



CPMAI Certification



Subscribe to Cognilytica newsletter on LinkedIn



Properly Scoping AI Projects [AI Today Podcast]



Prompt Engineering Best Practices: What is Prompt Chaining? [AI Today Podcast]



Prompt Engineering Best Practices: Using Custom Instructions [AI Today Podcast]


Prompt Engineering Best Practices: Hack and Track [AI Today Podcast]

Plugins for Large Language Models (LLMs) are additional tools or extensions that enhance the LLM's capabilities beyond its base functions. In this episode hosts Kathleen Walch and Ron Schmelzer discuss this topic in greater detail.



Can I use plugins with ChatGPT?



Plugins can access external databases, perform specific computations, or interact with other software and APIs to fetch real-time data, execute code, and more. In essence, they significantly expand the utility of LLMs, making them more versatile and effective tools for a wide range of applications. They bridge the gap between the static knowledge of a trained model and the dynamic, ever-changing information and capabilities of the external world. Plugins can be used on many different LLMs.



Why use plugins?



People use plugins for a variety of reasons. They allow you access to Real-time information by accessing up-to-date information from the web or other data sources,. They can also an perform specialized tasks like solving complex mathematical problems, generating code, or providing translations with expertise that might not be fully developed in the base model. Plugins also enable LLMs to interact with other applications and services, allowing for dynamic content generation, automation of tasks, and enhanced user interactions. They also allow for customization and personalization as well as improved performance and efficiency. In the episode we discuss this all in greater detail.



Show Notes:




Free Intro to CPMAI course



CPMAI Certification



Subscribe to Cognilytica newsletter on LinkedIn



Properly Scoping AI Projects [AI Today Podcast]



Prompt Engineering Best Practices: What is Prompt Chaining? [AI Today Podcast]



Prompt Engineering Best Practices: Using Custom Instructions [AI Today Podcast]


Prompt Engineering Best Practices: Using Plugins [AI Today Podcast]

As folks continue to use LLMs, best practices are emerging to help users get the most out of LLMs. OpenAI's ChatGPT allows users to tailor responses to match their tone and desired output goals. Many have reported that using custom instructions results in much more accurate, precise, consistent, and predictable results. But why would you want to do this and why does it matter? In this episode, hosts Kathleen Walch and Ron Schmelzer discuss why this is a best practice.



What are custom instructions in ChatGPT?



In ChatGPT, custom instructions are provided by answering two questions asked in settings that get sent along with your prompts:




What would you like ChatGPT to know about you to provide better responses?



How would you like ChatGPT to respond?




It's important to note that once created, these instructions will apply to all future chat prompt sessions (not previous or existing ones). This allows you to make somewhat permanent settings that don’t have to be constantly reset. Custom prompt instructions are short and generally limited to about 1500 characters, so keep it precise and concise. 



Show Notes:




Free Intro to CPMAI course



CPMAI Certification



Subscribe to Cognilytica newsletter on LinkedIn



Properly Scoping AI Projects [AI Today Podcast]



Prompt Engineering Best Practices: What is Prompt Chaining? [AI Today Podcast]



AI Today Podcast: AI Glossary Series – OpenAI, GPT, DALL-E, Stable Diffusion



AI Today Podcast: AI Glossary Series – Tokenization and Vectorization


Prompt Engineering Best Practices: Using Custom Instructions [AI Today Podcast]

Companies of all sizes in every  industry are looking to see how Artificial Intelligence (AI), machine learning (ML), and cognitive technology projects can provide them a competitive edge. They want to provide efficiencies and improve ROI in today’s competitive landscape. As a result, this creates tremendous opportunity in the field of AI for professionals who are CPMAI certified and follow the CPMAI methodology. This globally recognized certification helps you stand out from the pack when it comes to scoping, managing, and understanding AI and ML projects. In this episode of the AI Today podcast we interview George Fountain. He is Senior Project Manager at Booz Allen Hamilton (BAH), and is CPMAI certified.



What is CPMAI?



Staying up to date on AI in today's rapidly changing world is critical. However, just reading the news about AI versus actually being able to apply AI best practices to projects are two different things. As a senior project manager, George needs to have a step by step framework to successfully run and manage client projects. With the recent buzz around GenAI this has now become more important than ever. CPMAI is providing the framework needed. George shares how he is applying CPMAI in the real world. 



Is the CPMAI certification worth it?



When it comes to investing your time and money into certifications you want to make sure it's worth it. For George, CPMAI has been a transformational certification to his career. It's helped him grow key skills including providing the knowledge, lexicon, and credibility he was after. George shares his feedback and experience on this podcast. He wraps up the discussion by sharing how he believes AI will impact organizations in the years ahead.



Show Notes:




CPMAI Training and Certification



Cognilytica newsletter



Applying CPMAI Methodology to AI Projects: Interview with Christina Kucek, CAI [AI Today Podcast]


Applying CPMAI Methodology in the real world: Interview with George Fountain, Booz Allen Hamilton (BAH) [AI Today Podcast]

To improve the reliability and performance of LLMs, sometimes you need to break large tasks/prompts into sub-tasks. Prompt chaining is when a task is split into sub-tasks with the idea to create a chain of prompt operations. Prompt chaining is useful if the LLM is struggling to complete your larger complex task in one step. In this episode of the AI Today podcast hosts Kathleen Walch and Ron Schmelzer discuss prompt chaining. This is part 2 in our 6 part series on Prompt Engineering Best Practices.



Why do you need to chain the prompts?



In this episode we explain when and why to use prompts. Chaining prompts are great, and is a best practices approach, when you're answering questions from information in a document. It's also good when you want to validate & refine Responses. Or, when you're looking to Simplify Writing long-form content and break the writing process into outlined sections or chapters that the AI can expand upon in sequence. Prompt chaining is also good for providing a stepwise approach to research projects. And, for anything that needs an iterative approach such as iterative data analysis, iterative Approach to Computer programming, and iterative approach to other process-based tasks.



What is the prompt chaining technique?



There are a few different types of chain prompts. We go over Chain-of-thought prompting which is breaking down big tasks. We also discuss Self-consistency and ReAct techniques which is thinking harder and smarter. And, we discuss Flipped Interaction as well as Question Refinement Pattern.



When looking to chain prompts there are some best practices to follow. These best practices include decomposing the task, crafting and testing your prompts, chaining prompts, and evaluate and iterate as needed. Your prompt won't be perfect on the first try. Like we always say: Think Big, Start Small, and Iterate Often.



This is part 2 in our 6 part series on Prompt Engineering Best Practices. Subscribe to AI Today to get notified of upcoming episodes in this series and learn the best practices for prompt engineering from Cognilytica AI thought leaders.



Show Notes:




Free Intro to CPMAI course



CPMAI Certification



Subscribe to Cognilytica newsletter on LinkedIn



Properly Scoping AI Projects [AI Today Podcast]


AI Today Podcast: Generative AI Series: Implementing Generative AI in production

Download our free app to listen on your phone