DEV Community: Amit Sarkar

Building the Foundation: Your Pathway to Mastery in Prompt Engineering.

Amit Sarkar — Mon, 08 Jul 2024 17:51:27 +0000

Building the Foundation: Your Pathway to Mastery in Prompt Engineering

Greetings, courageous adventurer! You are about to embark on an intriguing journey through the domain of AI prompt engineering, poised to
learn, experiment, and innovate. This series serves as your compass, a powerful guide that will navigate you from novice to adept. Remember, each stride on this journey is a steppingstone towards mastering prompt engineering, a skill that you are absolutely capable of acquiring.

Though some of the initial codes are from my book on prompt engineering (revised edition will be released by the end of the month), the basic concepts of prompt engineering remain the same.
We will apply these principles while using Bedrock and the foundation models to develop real-life products. Follow along with me here and other places where I will share my own product development experience.

I am on a mission to build a tribe of AI product developers so that you rule the job market, not the other way around. A sure path to get a job fast if you follow along with me. I believe in you, and I am thrilled to walk alongside you on this exhilarating journey. Your fundamentals will be solid, your knowledge expansive, and your future as a prompt engineering expert bright. Here's to the journey ahead, and the mastery that awaits at the end!

Here's a sneak peek at the treasures awaiting you in this journey!

1.Super-Basic Prompts: Just like a baby's first words, we will start with the simple, foundational aspects of AI prompts.
2.The Recipe for a Prompt: Discover the ingredients of a successful prompt and how they blend together.
3.Placing Instructions at the Forefront of the Prompt: Understand the importance of clear, upfront instructions for garnering desired responses.
4.Prompts Aimed at Extracting Specific Info: Unleash the power of targeted prompts for precision extraction of information.
5.Zero-Shot Prompting: Delve into the mysterious world of zero-shot prompting and understand its intricacies.
6.One-Shot Prompting: Take a step further into one-shot prompting, witnessing the power of example-driven interactions.
7.Few-Shot Prompting: Traverse the nuanced landscape of few-shot prompting, where multiple examples guide your AI.
8.Setting a Role for ChatGPT: Play director in your own AI movie, setting roles and dictating performances.
9.Encouraging Questions in a Prompt: Discover how to spark curiosity within the AI model, encouraging a more dynamic exchange.
10.Explicit Prompts: Master the art of directness, commanding your AI to provide the exact response you seek.
11.Ambiguous Scenarios: Learn to navigate the foggy waters of ambiguity, drawing clarity from the AI's understanding.
12.Emulating a Dialogue: Step into the theatre of conversation, learning to orchestrate engaging back-and-forth with your AI.
13.Setting Your Own Role in a Prompt: Put yourself into the script, enhancing the personalization and relevance of the AI's responses.
14.And now, for the grand finale - an impressive array of 40 practice assignments, carefully distributed with 5 at the conclusion of each topic. This deliberate arrangement ensures that you thoroughly comprehend and master each topic before progr
essing to the next. These exercises are purposefully designed to cement your burgeoning skills, setting you on the path to true proficiency in the art of prompt engineering.
Feel the buzz of anticipation yet? Let's forge ahead, for this journey is one of discovery, transformation, and immense learning. You're about to become an expert, and we can't wait to see the magic you will create. Onward, prompt engineer!

Innovate! Don't Waste Your Skills. Think Like Wright Brothers!

Amit Sarkar — Fri, 05 Jul 2024 23:08:01 +0000

*Prioritizing engineering reasoning and critical thinking, akin to the Wright Brothers’ innovative mindset, is crucial for executives and technical managers to drive company-wide innovation. These skills enable the anticipation of trends, the solving of complex problems, and the fostering of a creative, problem-solving culture. By embedding these principles at the core of a company’s ethos, it not only leads to groundbreaking advancements but also secures a competitive market position. *

Learn AI The Best Way Bite Size

Amit Sarkar — Fri, 05 Jul 2024 18:37:50 +0000

Learn AI From My Own Product Development Experience.If You Do That The Entire Silicon Valley Will Open For You. I will post technical articles here and will give you everything you need. I will share my entire git repo with you. Prerequiste: Python Proficiency. I can give you a free book in the future. I am too busy for that now.

What you will learn.

Prompt Engineering - The real Deep Dive not any Kindle book vodoo " Learn 100 prompts that will make you rich" You have seen such hypes all over. If it is that simple, trust me I would be flying in my private jet by now.
Lang Chain
Vector Database. Pinecone as well as MongoDB Atlas
Amazon Bedrock
Amazon Q
Amazon CodeWhisperer
Amazon Lambda

Once you learn and master all these then comes how you can build a product. I will share my real life experience on that. Wait Dude, not so fast. You need to have engineering reasoning and critical thinking skills to be an innovator. So, watch this first. I know you are busy but spend next 5 minutes and watch this for an important message. [https://www.youtube.com/watch?v=_7hIqf7skiQ]

Here is your first lesson on prompt engineering. You can run this on Google Colab or any Jupyter Notebook.

!pip install openai

from openai import OpenAI
client = OpenAI(
api_key = 'your openai apikey',
)
**

Define the prompt for the language model

prompt = 'Write an essay about the History of the Silicon Valley and how this place literally changed the world'

# Call the OpenAI API with the specified model, prompt, maximum token length, and temperature

output = client.completions.create(
model="gpt-3.5-turbo-instruct",
prompt=prompt,
max_tokens=1000,
temperature=0.5 )

# Print the generated text
print(output.choices[0].text)
**
Here is a line-by-line explanation of the code:**

“import openai” - This line imports the OpenAI library into your Python environment. This library allows our program to interact with OpenAI's APIs, which include language models like GPT-3.
“secret_key” = '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'- This line sets your OpenAI API secret key to a variable namedsecret_key`. You will need to replace this with your actual secret key. I have used my secret_key here to show you how to use it. I will delete this secret key as soon I publish this here.
“openai.api_key” = secret_key - This line assigns the secret key to the OpenAI API key. This authenticates your application with OpenAI's servers.
“prompt” = 'Write an essay about the History of the Silicon Valley and how this place literally changed the world' - This line sets the variable 'prompt to the text that will be used to prompt the language model.
“output” = openai.Completion.create( )- This line starts the call to OpenAI's API to create a new text completion. The result of this call will be stored in the output variable. openai.Completion.create() function generates text based on the provided prompt using the specified engine.
“model”="gpt-3.5-turbo-instruct",` - This specifies the language model to be used for the text generation task. In this case, the "gpt-3.5-turbo-instruct" model is being used.
“prompt=prompt” - This feeds in the prompt variable defined earlier to the language model.
“max_tokens=1000” - This line tells the API to limit the response to 1000 tokens (a token can be as short as one character or as long as one word).
“temperature=0.5” - This sets the randomness of the output. Lower values (closer to 0) make the output more deterministic and focused, while higher values (closer to 1) produce more diverse and creative outputs.
“print(output.choices[0].text)” - This line prints the generated text. The API returns a list of choices (though by default it is only one choice). This command prints the text field of the first choice.

By running this single program, you have achieved many critical tasks. You learned to:
• Import the OpenAI API library
• Assign your API key to a variable
• Set the OpenAI API key for authentication
• Define the prompt for the language model
• Call the OpenAI API with the specified model, prompt, maximum
token length, and temperature
• Print the generated text
This is a monumental accomplishment!

*Please make your comments. I need your feedback. Stay here for regular new code uploads. Once you master this you can join my tribe of start up entrepreneurs or get a job at Silicon Valley. Who knows! AWS might hire you. *

Do you know the famous motivational speaker Tony Robins always says "Live with passion". Hey Tony, with all due respect dude you are wrong! My mantra is "You will have to be obsessed with something to become unstoppable"

I am obsessed with AI and AWS as well. Are you?

Learn Generative AI from Real Product Development Experience

Amit Sarkar — Tue, 02 Jul 2024 19:16:19 +0000

Learn Generative AI from Real Product Development Experience. Yes, you heard that right! https://www.meetup.com/meetup-group-komabzoz/events/301904960/
Join us every 2nd Thursday of the month to learn Generative AI from actual product development experience. I will share my insights and experiences working on two Generative AI products on the AWS platform using Anthropic, Vector Database, and several other services.
The first product is scheduled to go live by the end of July 2024. This series of hands-on workshops will cover:

Generative AI
Prompt Engineering
Vector Database
Building Software with Generative AI
Generative AI and the Future of Work Come and join me in this jaw-dropping hands-on session on Generative AI. Each month, you'll learn something new and get closer to developing your own product or starting an exciting career in the rapidly growing field of Generative AI. Don't miss out on this incredible opportunity to dive into the world of Generative AI!