Close Menu
    Trending
    • Paper Insights: Masked Autoencoders that listen | by Shanmuka Sadhu | Jun, 2025
    • Statistical Inference: Your Friendly Guide to Making Sense of Data | by Timothy Kimutai | Jun, 2025
    • Machine Learning in Finance: Next-Gen Budget Forecasting | by Kavika Roy | Jun, 2025
    • Discover How AI Can Transform the Way You Work With This $20 E-Degree
    • When Your Probabilities Lie — A Hands-On Guide to Probability Calibration | by Anirban Mukherjee | Jun, 2025
    • Why Qualitative Feedback Is the Most Valuable Metric You’re Not Tracking
    • ChatGPT-4.5: OpenAI’s Most Powerful AI Model Yet! | by Sourabh Joshi | Jun, 2025
    • Building Wealth While Building a Business: 10 Financial Habits That Pay Off Long-Term
    Finance StarGate
    • Home
    • Artificial Intelligence
    • AI Technology
    • Data Science
    • Machine Learning
    • Finance
    • Passive Income
    Finance StarGate
    Home»Artificial Intelligence»How Would I Learn to Code with ChatGPT if I Had to Start Again
    Artificial Intelligence

    How Would I Learn to Code with ChatGPT if I Had to Start Again

    FinanceStarGateBy FinanceStarGateMay 1, 2025No Comments11 Mins Read
    Share Facebook Twitter Pinterest LinkedIn Tumblr Reddit Telegram Email
    Share
    Facebook Twitter LinkedIn Pinterest Email


    Coding has been part of my life since I used to be 10. From modifying HTML & CSS for my Friendster profile in the course of the easy web days to exploring SQL injections for the joys, constructing a three-legged robotic for enjoyable, and recently diving into Python coding, my coding journey has been numerous and enjoyable!

    Right here’s what I’ve realized from numerous programming approaches.

    The way in which I be taught coding is all the time comparable; As individuals say, principally it’s simply copy-pasting. 😅

    With regards to constructing one thing within the coding world, right here’s a breakdown of my methodology:

    1. Select the Proper Framework or Library
    2. Be taught from Previous Tasks
    3. Break It Down into Steps
      Slice your venture into actionable merchandise steps, making growth much less overwhelming.
    4. Google Every Chunk
      For each step, seek the advice of Google/Bing/DuckDuckGo/any search engine you favor for insights, steering, and potential options.
    5. Begin Coding
      Attempt to implement every step systematically.

    Nonetheless, even essentially the most well-thought-out code can encounter bugs. Right here’s my technique for troubleshooting:

    1. Verify Framework Documentation: ALWAYS learn the docs!

    2. Google and Stack Overflow Search: search on Google and Stack Overflow. Instance key phrase could be:

    web site:stackoverflow.com [coding language] [library] error [error message]

    web site:stackoverflow.com python error ImportError: pandas module not discovered

    – Stack Overflow Options: If the problem is already on Stack Overflow, I search for essentially the most upvoted feedback and options, typically discovering a fast and dependable reply.
    – Belief My Instinct: When Stack Overflow doesn’t have the reply, I belief my instinct to seek for reliable sources on Google; GeeksForGeeks, Kaggle, W3School, and In the direction of Data Science for DS stuff 😉

    3. Copy-Paste the Code Resolution

    4. Confirm and Check: The ultimate step consists of checking the modified code totally and testing it to make sure it runs as meant.

    And Voila you simply resolve the bug!

    Photograph by Stephen Hocking on Unsplash

    Isn’t it stunning?

    However in actuality, are we nonetheless doing this?!

    Recently, I’ve observed a shift in how new coders are tackling coding. I’ve been instructing how one can code professionally for about three years now, bouncing round in coding boot camps and visitor lecturing at universities and company coaching. The way in which coders are stepping into code studying has modified a bit.

    I normally inform the recent faces to stay with the old-school methodology of searching and googling for solutions, however persons are nonetheless utilizing ChatGPT ultimately. And their alibi is

    “Having ChatGPT (for coding) is like having an additional research buddy -who chats with you want a daily particular person”.

    It is useful, particularly if you’re nonetheless making an attempt to wrap your head round issues from search outcomes and documentation — to develop what’s so-called programmer instinct.

    Now, don’t get me fallacious, I’m all for the fundamentals. Shopping, studying docs, and throwing questions into the neighborhood pot — these are stable strikes, in my guide. Relying solely on ChatGPT is perhaps a bit a lot. Certain, it could actually whip up a speedy abstract of solutions, however the standard searching strategies provide the freedom to choose and select, to experiment a bit, which is fairly essential within the coding world.

    However, I’ve gotta give credit score the place it’s due — ChatGPT is lightning-fast at giving out solutions, particularly if you’re nonetheless making an attempt to determine the correct from the fallacious in search outcomes and docs.

    I notice this shift of utilizing ChatGPT as a research buddy is just not solely taking place within the coding scene, Chatgpt has revolutionized the way in which individuals be taught, I even use ChatGPT to repair my grammar for this put up, sorry Grammarly.

    Saying no to ChatGPT is like saying no to search engines like google and yahoo within the early 2000 period. Whereas ChatGPT might include biases and hallucinations, just like search engines like google and yahoo having unreliable data or hoaxes. When ChatGPT is used appropriately, it could actually expedite the training course of.

    Now, let’s think about a real-life state of affairs the place ChatGPT may aid you by being your coding buddy to assist with debugging.

    State of affairs: Debugging a Python Script

    Think about you’re engaged on a Python script for a venture, and also you encounter an sudden error that you would be able to’t resolve.

    Right here is how I was taught to do it — the period earlier than ChatGPT.

    Shopping Method:

    1. Verify the Documentation:

    Begin by checking the Python documentation for the module or perform inflicting the error.

    For instance:
    – go to https://scikit-learn.org/stable/modules/ for Scikit Be taught Doc

    2. Search on Google & Stack Overflow:

    If the documentation doesn’t present an answer, you flip to Google and Stack Overflow. Scan via numerous discussion board threads and discussions to discover a comparable difficulty and its decision.

    StackOverflow Thread

    3. Belief Your Instinct:

    If the problem is exclusive or not well-documented, belief your instinct! You may discover articles and sources on Google that you simply’ve discovered reliable up to now, and attempt to adapt comparable options to your downside.

    Google Search Outcome

    You may see that on the search end result above, the outcomes are from W3school – (trusted coding tutorial web site, nice for cheatsheet) and the opposite 2 outcomes are official Pandas documentation. You may see that search engines like google and yahoo do counsel customers have a look at the official documentation. 😉

    And that is how you should utilize ChatGPT that can assist you debug a difficulty.

    New Method with ChatGPT:

    1. Interact ChatGPT in Conversations:

    As an alternative of solely navigating via documentation and boards, you may have interaction ChatGPT in a dialog. Present a concise description of the error and ask. For instance,

    “I’m encountering a difficulty in my [programming language] script the place [describe the error]. Are you able to assist me perceive what is perhaps inflicting this and counsel a doable answer?”

    Interact ChatGPT in Conversations

    2. Make clear Ideas with ChatGPT:

    If the error is said to an idea you’re struggling to know, you may ask ChatGPT to elucidate that idea. For instance,

    “Clarify how [specific concept] works in [programming language]? I believe it is perhaps associated to the error I’m going through. The error is: [the error]”

    Make clear Ideas with ChatGPT

    3. Search Suggestions for Troubleshooting:

    You ask ChatGPT for common tips about troubleshooting Python scripts. For example,

    “What are some frequent methods for coping with [issue]? Any suggestions on instruments or methods?”

    Utilizing ChatGPT as coding buddy

    Potential Benefits:

    • Customized Steerage: ChatGPT can present personalised steering based mostly on the particular particulars you present concerning the error and your understanding of the issue.
    • Idea Clarification: You may search explanations and clarifications on ideas immediately from ChatGPT leveraging their LLM functionality.
    • Environment friendly Troubleshooting: ChatGPT may supply concise and related ideas for troubleshooting, doubtlessly streamlining the debugging course of.

    Potential Limitations:

    Now let’s discuss concerning the cons of counting on ChatGPT 100%. I noticed these points so much in my pupil’s journey on utilizing ChatGPT. Put up ChatGPT period, my college students simply copied and pasted the 1-line error message from their Command Line Interface regardless of the error being 100 strains and linked to some modules and dependencies. Asking ChatGPT to elucidate the workaround by offering a 1 line error code may work generally, or worse — it would add 1–2 hour manhour of debugging.

    ChatGPT comes with a limitation of not with the ability to see the context of your code. For certain, you may all the time give a context of your code. On a extra advanced code, you won’t be capable to give each line of code to ChatGPT. The truth that Chat GPT solely sees the small portion of your code, ChatGPT will both assume the remainder of the code based mostly on its information base or hallucinate.

    These are the doable limitations of utilizing ChatGPT:

    • Lack of Actual-Time Dynamic Interplay: Whereas ChatGPT gives worthwhile insights, it lacks the real-time interplay and dynamic back-and-forth that boards or dialogue threads may supply. On StackOverflow, you may need 10 totally different individuals who would counsel 3 totally different options which you’ll examine both by DIY ( do it your self, attempt it out) or see the variety of upvotes.
    • Dependence on Previous Information: The standard of ChatGPT’s response relies on the knowledge it has been skilled on, and it will not be conscious of the most recent framework updates or particular particulars of your venture.
    • May add further Debugging Time: ChatGPT doesn’t have a context of your full code, so it would lead you to extra debugging time.
    • Restricted Understanding of Idea: The standard searching strategies provide the freedom to choose and select, to experiment a bit, which is fairly essential within the coding world. If you understand how to handpick the correct supply, you in all probability be taught extra from searching by yourself than counting on the ChatGPT common mannequin.
      Except you ask a language mannequin that’s skilled and specialised in coding and tech ideas, analysis papers on coding supplies, or well-known deep studying lectures from Andrew Ng, Yann Le Cunn’s tweet on X (previously Twitter), just about ChatGPT would simply give a common reply.

    This state of affairs showcases how ChatGPT could be a worthwhile device in your coding toolkit, particularly for acquiring personalised steering and clarifying ideas. Bear in mind to stability ChatGPT’s help with the strategies of searching and ask the neighborhood, preserving in thoughts its strengths and limitations.


    Ultimate Ideas

    Issues I’d suggest for a coder

    For those who actually wish to leverage the autocompletion mannequin; as a substitute of solely utilizing ChatGPT, attempt utilizing VScode extensions for auto code-completion duties comparable to CodeGPT — GPT4 extension on VScode, GitHub Copilot, or Google Colab Autocomplete AI instruments in Google Colab.

    Auto code completion on Google Colab

    As you may see within the screenshot above, Google Colab mechanically offers the consumer recommendations on what code comes subsequent.

    One other various is Github Copilot. With GitHub Copilot, you will get an AI-based suggestion in real-time. GitHub Copilot suggests code completions as builders sort and switch prompts into coding recommendations based mostly on the venture’s context and elegance conventions. As per this release from Github, Copilot Chat is now powered by OpenAI GPT-4 (a similiar mannequin that ChatGPT is utilizing).

    Github Copilot Instance — picture by Github

    I’ve been actively utilizing CodeGPT as a VSCode Extension earlier than I knew that Github Copilot is accessible at no cost if you’re in schooling program. CodeGPT Co has 1M obtain to this date on the VSCode Extension Market. CodeGPT permits seamless integration with the ChatGPT API, Google PaLM 2, and Meta Llama.
    You will get code recommendations via feedback, right here is how:

    • Write a remark asking for a particular code
    • Press cmd + shift + i
    • Use the code 😎

    You may as well provoke a chat through the extension within the menu and leap into coding conversations 💬

    As I replicate on my coding journey, the invaluable lesson realized is that there’s no one-size-fits-all strategy to studying. It’s important to embrace a various array of studying strategies, seamlessly mixing conventional practices like searching and neighborhood interplay with the modern capabilities of instruments like ChatGPT and auto code-completion instruments.

    What to Do:

    • Make the most of Tailor-made Studying Sources: Profit from ChatGPT’s suggestions for studying supplies.
    • Collaborate for Drawback-Fixing: Make the most of ChatGPT as a collaborative associate as if you’re coding with your pals.

    What To not Do:

    • Over-Dependence on ChatGPT: Keep away from relying solely on ChatGPT and guarantee a balanced strategy to foster unbiased problem-solving abilities.
    • Neglect Actual-Time Interplay with Coding Group: Whereas ChatGPT gives worthwhile insights, don’t neglect the advantages of real-time interplay and suggestions from coding communities. That additionally helps construct a repute locally
    • Disregard Sensible Coding Observe: Steadiness ChatGPT steering with hands-on coding follow to bolster theoretical information with sensible software.

    Let me know within the feedback how you employ ChatGPT that can assist you code!
    Comfortable coding!
    Ellen

    🌐 Observe me on LinkedIn
    🚀 Take a look at my portfolio: liviaellen.com/portfolio
    👏 My Earlier AR Works: liviaellen.com/ar-profile
    ☕ or simply buy me a real coffee ❤ — Sure I like espresso.

    Concerning the Creator

    I’m Ellen, a Machine Learning engineer with 6 years of expertise, at the moment working at a fintech startup in San Francisco. My background spans information science roles in oil & gasoline consulting, in addition to main AI and information coaching applications throughout APAC, the Center East, and Europe.

    I’m at the moment finishing my Grasp’s in Knowledge Science (graduating Could 2025) and actively searching for my subsequent alternative as a machine studying engineer. For those who’re open to referring or connecting, I’d actually respect it!

    I like creating real-world impression via AI and I’m all the time open to project-based collaborations as nicely.



    Source link

    Share. Facebook Twitter Pinterest LinkedIn Tumblr Email
    Previous ArticleΠώς λειτουργεί το advantage+ audience της Meta; | by Knowcrunch | May, 2025
    Next Article AI Is Taking Over Coding at Microsoft, Google, and Meta
    FinanceStarGate

    Related Posts

    Artificial Intelligence

    How AI Agents “Talk” to Each Other

    June 14, 2025
    Artificial Intelligence

    Stop Building AI Platforms | Towards Data Science

    June 14, 2025
    Artificial Intelligence

    What If I had AI in 2018: Rent the Runway Fulfillment Center Optimization

    June 14, 2025
    Add A Comment

    Comments are closed.

    Top Posts

    What’s Your Hacker Name? Tale of Weak passwords | by Zeeshan Saghir | Apr, 2025

    April 3, 2025

    Virtual Medical Scribe Solution: Best Practices for Remote Teams

    April 17, 2025

    This is How So Many Business Owners Learned to Use AI for Automation

    April 27, 2025

    What First Names Are the Most Successful in Business?

    February 28, 2025

    Trying Your Best When There’s Little To No Chance Of Succeeding

    February 26, 2025
    Categories
    • AI Technology
    • Artificial Intelligence
    • Data Science
    • Finance
    • Machine Learning
    • Passive Income
    Most Popular

    LSTM’s. Welcome to ML Decoded, where I share my… | by Biren Mer | Mar, 2025

    March 4, 2025

    The Dangers of Deceptive Data Part 2–Base Proportions and Bad Statistics

    May 9, 2025

    Reinforcement Learning Made Simple: Build a Q-Learning Agent in Python

    May 27, 2025
    Our Picks

    Creating an AI Agent to Write Blog Posts with CrewAI

    April 4, 2025

    The True Power of AI and Its Impact on the World | by Oyakhiredarlington | Feb, 2025

    February 12, 2025

    The Future of Work: Navigating the Evolution of the Job Market Amid AGI Advancement | by Deepan Ilangkamban Ramesh | Jun, 2025

    June 2, 2025
    Categories
    • AI Technology
    • Artificial Intelligence
    • Data Science
    • Finance
    • Machine Learning
    • Passive Income
    • Privacy Policy
    • Disclaimer
    • Terms and Conditions
    • About us
    • Contact us
    Copyright © 2025 Financestargate.com All Rights Reserved.

    Type above and press Enter to search. Press Esc to cancel.