WHAT IS PROMPT ENGINEERING ?

WHAT IS PROMT ENGINEERING ???

-Prompt engineering is a concept in artificial intelligence (AI), particularly natural language processing (NLP). In prompt engineering.

- Prompt engineering refers to the process of designing and developing high-quality prompts for natural language processing (NLP) models.

In the context of machine learning and AI, prompts are the textual or verbal cues provided to an algorithm to initiate a specific task or generate

a particular output.

 Prompt Engineering, also known as In-Context Prompting, refers to methods for how to communicate with LLM to steer its behavior for desired outcomes without

updating the model weights. 

 It is an empirical science and the effect of prompt engineering methods can vary a lot among models, thus requiring heavy experimentation and heuristics.

Many useful resources...

-AI BOOK many in-depth examples for how to utilize LLM efficiently.

-A library for combining language models with other components to build applications.

-Prompt Engineering Guide repo contains a pretty comprehensive collection of education materials on prompt engineering.

-learnprompting.org

-PromptPerfect

-Semantic Kernel

 - A Prompt guides the model to generate usefull outputs. try to multiple formulations of your prompt to get the best generations.

describe the task and general setting

   -there are so many different kinds of generative AI models you can think about a image generator model right like where you put

certain kinds of caption and it is able to generate the exact kind of image output just out of thin air right you would have seen a lot of people putting

out pictures that they generated from dally things like a dog wearing space suit and flying in the air or um an elephant carrying an ice cream

right things like that so these are some of the specific prompts that you're gonna give to that model so it's able to generate the very specific results that.

  you're looking for now you might be wondering why is it important and if you have ever done this in the past a lot of people also said that prompt

engineering is something which existed before and it wasn't until the generative AI models came into picture that it was tagged as prompt engineering

so I do believe the same and I'll give an example that I felt is very similar to what we are calling prompt engineering for the generative AI space

right.

 

   when you're using Google search I use a lot of tricks to generate very specific and optimized search results so for example if I want specific keywords

to appear in my search results I use quotation marks that makes sure that each and every result that I'm seeing on my first paid second page Etc contains

that particular keyboard if I want a particular keyword to not appear in my search results then I use a minus sign and put that keyword in there so that

that particular keyword does not appear in my search results.

   if I want a particular regular file format let's say I'm looking for a Time series presentation then I would say time series PPT or time series PDF and it

will generate the search results which contains those file types so what are these things right these are prompt engineering that you're doing with

Google search similarly when you are using a app like chat GPT or any other generative AI system how do you optimize the results.

  that you want is what is called as prompt engineering now a lot of you were asking how do you learn from the engineering there is no one source

that will be helpful for you to learn the art of prompt engineering because there's a lot of components that go into prompt engineering you need to

understand the capabilities of that particular system in. 

  and out which means what are the kind of results this system can produce what are the kind of output formats that it can produce what is the

extent of knowledge that this system has what is the kind of questions or likelanguage components that it can understand.

 

   I'll give you an example if you go to chat GPT and you ask it hey give me a diagram of ml Ops pipeline it's not going to give that right it's going to

say that hey I'm uh I'm a language trained model and I only produce text-based results so you need to understand that the knowledge breadth or

like the output Brits that chart GPT has is restricted to text output similarly.

  uh it's it's said that chat GPT is trained Based on data up till 2021 which means that any question that you're going to ask to it which dates after

that it's not going to be able to answer similarly there are restrictions or subject matter.

  are interested people in learning prompt engineering there are a few resources which I'm going to put down in and one book which you

might want to read is the gpt3 by hugging face which I found was very helpful there's also another book called Transformers for natural language

processing uh which is published by Pact which I also found was very helpful so you can read that these books are not going to be specifically talking about.

  prompt engineering but what they'll be doing is to make you understand that what are the components that go into building large language models how do

you tune these large language models how do you understand the capabilities of these large language models and how do you come up with minimal uh say like

sentences in English or any particular language so you're able to extract the best kind of outputs from these kind of models so hope that is helpful.

--JOBS ABOUT PROMPT ENGINEERING ....

 AI TOOLS to delevers more accurate ane relatively respones to the questions real people are like to the pose.

annual income of prompt engineering is a google says INR 3.6 LACS.

-=- A good Prompt Engineers must have strong problem-solving and analytical skills, as well as a deep understanding of AI,

 natural language processing, machine learning and AI-generated content development.

- there is a high demanding for these professions in industrial like customer service and health caring and financial.

 -and that only requires minimal coading experience. many non-coaders achieve success in this fields.

Just note that low barriers to entry create a competitive job market—broaden your options by creating prompts for different LLMs, large language models.