Introduction

Let's establish the tone right away.

I am a lazy developer who is constantly looking for ways to automate mundane coding tasks.

Well, since Chat-GPT was trained on a massive corpus of data from genius developers, why not use that to help me become lazier?

I admit I had my doubts about AI chatbots.

Just look at Wysa!

From Boilerplate to Complex Code in Seconds

Gone are the days of spending hours writing boilerplate code.

Chat-GPT generates it for me in a matter of seconds.

• Need a function that performs a specific task?

  • ✅ Chat-GPT

• Need to generate code snippets for complex algorithms and data structures?

  • ✅ Chat-GPT

• Stuck on a particular problem or can't think of the best way to implement a solution?

  • ✅ Chat-GPT

• Do my laundry.

  • ❌ Chat-GPT...not yet buddy.

Quality or Quantity?

How good is the code that Chat-GPT generates, though?

There are times when the code it generates is of high quality and follows excellent coding principles.

At other times, it would generate code that was a little buggy or didn't quite do what I needed it to do.

Of course, the better the prompts, the "better" the results.

I'm placing a HUGE caveat on better, as it is generally harder to read and comprehend overly concise or dense code than one that is more verbose.

"""
    dict1 = {'a': 1, 'b': 2, 'c': 3}
    dict2 = {'a': 4, 'b': 5, 'd': 6}
"""
def group_similar_keys(dict1, dict2):
    common_keys = set(dict1.keys()).intersection(dict2.keys())
    result_dict = {}
    for key in common_keys:
        result_dict[key] = [dict1[key], dict2[key]]
    return result_dict

"""
    dict1 = {'a': 1, 'b': 2, 'c': 3}
    dict2 = {'a': 4, 'b': 5, 'd': 6}
"""
def group_similar_keys(dict1, dict2):
    result_dict = {key: [dict1[key], dict2[key]] for key in dict2.keys() if key in dict1}
    return result_dict

"""
    dict1 = {'a': 1, 'b': 2, 'c': 3}
    dict2 = {'a': 4, 'b': 5, 'd': 6}
"""
def group_similar_keys(dict1, dict2):
    return {key: [dict1[key], dict2[key]] for key in dict2.keys() if key in dict1}

import unittest

class TestGroupSimilarKeys(unittest.TestCase):
    
    def test_group_similar_keys(self):
        dict1 = {'a': 1, 'b': 2, 'c': 3}
        dict2 = {'a': 'x', 'b': 'y', 'd': 'z'}
        expected_output = {'a': [1, 'x'], 'b': [2, 'y']}
        self.assertEqual(group_similar_keys(dict1, dict2), expected_output)
        
        dict1 = {'x': 1, 'y': 2, 'z': 3}
        dict2 = {'a': 'x', 'b': 'y', 'c': 'z'}
        expected_output = {}
        self.assertEqual(group_similar_keys(dict1, dict2), expected_output)
        
        dict1 = {'a': 1, 'b': 2, 'c': 3}
        dict2 = {'a': 'x', 'b': 'y', 'c': 'z'}
        expected_output = {'a': [1, 'x'], 'b': [2, 'y'], 'c': [3, 'z']}
        self.assertEqual(group_similar_keys(dict1, dict2), expected_output)
        
        dict1 = {}
        dict2 = {'a': 'x', 'b': 'y', 'c': 'z'}
        expected_output = {}
        self.assertEqual(group_similar_keys(dict1, dict2), expected_output)
        
        dict1 = {'a': 1, 'b': 2, 'c': 3}
        dict2 = {}
        expected_output = {}
        self.assertEqual(group_similar_keys(dict1, dict2), expected_output)
        
if __name__ == '__main__':
    unittest.main()

It is generally harder to read and understand overly concise or dense code than code that is more verbose.

My productivity has increased by 30–40% as a result of using Chat-GPT to scaffold baseline code, which I have found to be extremely helpful in streamlining my workflow.

And that will undoubtedly get a lot better with time.

Learning Made Fun

Perhaps the most enjoyable aspect of using Chat-GPT for coding was how much fun it was.

I was able to converse with an AI assistant who made the coding learning process more engaging and interactive.

And the best part?

I didn't have to scour through Stack Overflow for hours or wade through dense technical manuals.

In just a few seconds, Chat-GPT was able to provide me with the information I required.

It was like having a coding guru right at my fingertips!

On top of that, Chat-GPT was also able to assist me with problem-solving in novel and creative ways.

I constantly annoy my pair programming partner with my sloppy coding habits.

Not any more, thanks to Chat-GPT!

The Pros of AI Assistance...

As a developer, I know the struggle of searching for answers to coding problems.

One main draw of using Chat-GPT is that it allows me to quickly find answers without spending time searching through online forums or manuals.

And let's be honest, sometimes I'm just too lazy to search for answers on my own.

The fact that I can ask Chat-GPT questions using natural language without worrying about complicated syntax or commands makes it so much easier to obtain the information I require.

Another fantastic benefit is that it can help me learn new coding concepts that I may be unfamiliar with.

I can simply ask a question and get a clear and concise explanation.

This is especially helpful for self-taught engineers like myself who may not have access to formal coding education.

It's Not All Sunshine and Roses...

Hallucinations. Yes, Chat-GPT also hallucinates code in addition to text.

There have been instances where Chat-GPT confidently generated code that did not work or compile. (surprise, surprise)

"""
    Generate python code for the folowing problem.
    # Given an integer, return the integer with reversed digits.
    # Note: The integer could be either positive or negative.
"""
def reverse_integer(num):
    if num >= 0:
        return int(str(num)[::-1])
    else:
        return -int(str(num)[::-1])

Not to mention, there is a potential for overreliance on Chat-GPT because it is so quick and simple to use.

While it does a wonderful job of answering very specific coding questions, it may struggle to provide solutions that are relevant to your requirements. (But that could soon change with GPT-4's enhanced contextual capabilities!)

Is Your Job at Risk?

There is a growing concern that the emergence of artificial intelligence (AI) tools like Chat-GPT could ultimately eliminate the need for human programmers.

It sounds far-fetched but the fact is that since the launch of Chat-GPT, AI technology has been advancing at an insane breakneck pace.

No really, it seems like news regarding Large Language Models pops out every minute.

It's also understandably why people are spreading FUD about Chat-GPT replacing their jobs.

The reality is that AI will and can automate mundane tasks and gain insights that would be impossible to obtain through conventional methods.

However, if you already possess technical skills, I honestly believe that this puts you in a better position than the majority of individuals whose occupations will be radically changed by AI.

In part because you can obtain a better grasp of how AI works and how it evolves and improves over time.

There will always be a demand for humans in fields relating to technology and artificial intelligence.

In some ways, now may be the ideal time to learn how to develop with AI.

Some engineers told me that this was cheating.

It would be foolish, in my opinion, to not embrace this technology and tool.

Maybe AI will evolve to become a personal genius-level coding buddy, the one who never gets annoyed by my sloppy coding habits.

Perhaps someday I'll be able to live out my dream of working remotely while taking not one, but TWO six-month vacations a year, all thanks to the power of AI! 😎