Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
My programming skills leave something to be desired. I never stuck with books and tutorials long enough to create the kinds of apps and software I wanted to see. AI-powered chatbots powered by large language models have changed that. Now I can jot down my vision for something like a horror movie visual gallery or a new twist on a classic computer game and the chatbot will turn my plain English into hundreds of lines of code – and bring my vision to life.
This is called Atmosphere coding. For experienced programmers, it can make work go faster. But for non-programmers, it puts programming tools in our hands. Having the right mindset for the process can make the difference between a good or bad experience — even if it doesn’t achieve the results you originally hoped for. Sometimes, you’ll need to learn what you can and can’t use in programming as you go.
You can choose Gemini, ChatGPT, Claude, or another chatbot, and each will have their own features and quirks. The approach you take and how you prompt your chatbot is just as important as the model you choose, because it all depends on your instructions.
Here are five tips that can help you get started with effective programming.
Get to know your chatbot
The chatbot you use is important, although the type of form may be more important than its maker.
When you try Create the same application Using both the Gemini 2.5 Flash and Gemini 3 Pro, it became clear that a different approach was needed with the former to achieve similar outputs to the more advanced Pro model. This means more intention, specific instructions with prompts and from my experience, more manual work.
Since I was programming with Gemini 2.5 Flash, it often made the process more practical, which I wasn’t looking for. When the app is optimized, it will only provide icon for the section that was modified, leaving me to manually swap the icon or follow up with another prompt to make the entire icon available. In contrast, with Gemini 3 Pro, it will automatically save the entire code text.
Your mileage may vary, but if you are deciding between using a “fast” model or a “thinking” model for your biocoding project, you may need more skills and training to work with a fast model. If all you have available is a quick or less advanced form that asks you to make modifications to the code that you’re not comfortable with, you can ask the chatbot again to provide all the code (or set a rule to provide the entire body of code with modifications after each change is made).
From my experience using many chatbots (but not all): If you don’t mind using a bit of code, you’ll likely find success with most forms as long as you make clear, specific prompts. Non-programmers: Stick to more advanced “inference” or “reasoning” models if you can. I’ve found that thought models will do more of the heavy lifting for you.
Knowing your chatbot, its limitations and capabilities comes from experience, so it’s best to use it generally and ask questions.
Make your claims specific
If you have a detailed idea of the type of project you want to undertake, specificity is key.
Vibe programming all starts with what’s in your head. If you have a very clear idea, you can and should include everything you want to see in your project in your initial prompt. Make it comprehensive. The chatbot will create what it can create, and hopefully you can see your idea taking shape.
On the other hand, you can have a very loose idea of what you want the app to be and give the AI more leeway to control how the app works or how to achieve something you asked of it in a different way.
Repetition, repetition, repetition
Your dream app probably won’t be among the first or second prompts you give to your chatbot. You’ll probably spend more time asking for improvements than anything else.
Ask your chatbot for suggestions. If you have a problem with a part of your web app, whether it’s how it works or its design, just ask your chatbot. It will provide you with as many suggestions as you want. Since reactive programming is an iterative process, being asked to suggest five ways to improve your application can make a real difference.
Pay attention to the technical stuff, just a little
Non-programmers will need to know some technical things to make the application or project work well. The good thing is that you can use the chatbot as a resource.
Here are some things you’ll want to find out:
appearance: For most simple web applications, outputting code in HTML format is ideal for non-programmers. If you don’t want to deal with multiple files and folders, you can request a single HTML page. This can introduce limitations and potential “memory issues” in the chatbot context window if the file becomes large, but it is one of the most direct ways to get the output code to your browser for testing. If you’re not sure, ask your chatbot what the best format should be for your project and app launch process.
size: If you’re not sure about the capabilities of a chatbot, just ask. It will tell you that it won’t be able to create a new social network for you, but if you want a voice visualizer that channels the old Winamp days, you’re in luck. If your request is beyond the chatbot’s capabilities, ask it for alternative methods.
Check for errors: Sometimes your chatbot will provide broken code, so you’ll need to test again and again. What’s most important is to communicate what’s not working – or exactly how you want something to work. If you make a vague claim, don’t be surprised to receive vague results. Telling a chatbot “this isn’t working” is much less effective than detailing the specific errors you encounter during testing. The more specific your feedback, the more accurately the AI will respond, making the overall experience more efficient and enjoyable.
Have an open mind
Unless you have a very specific and strict goal for your bioprogramming project, keeping an open mind is essential. If you don’t know anything about programming, you probably don’t know all the possibilities available to you. If you’re not sure of your limits, it’s easy to underestimate and overestimate what you can do.
In an ideal world, everything you program will run smoothly, but errors are likely to occur. Whether it’s your chatbot forgetting to remember something, having technical limitations, or something in between, some things are going to break. If you’re lucky, the chatbot will recognize the problem and fix it, but sometimes it won’t be able to do so, and this is where an open mind comes in handy. Ask for alternative approaches – You may find an approach you like more than your original idea. You’ll never know what you don’t know unless you ask.
Don’t be afraid to start over
Like other creative processes, sometimes starting over is best. If you’ve gone through countless iterations and your app isn’t where you want it to be, consider starting over. This can be done completely from scratch or by taking the code from the first chat and using it as the basis for the new chat you start.
Although you can instruct your chatbot to start from scratch in the original chat, there may be a clean slate moving forward. Try again in a new conversation – so the AI doesn’t confuse things with your previous project. You may have identified some prompts that may have sent your project backwards that you can avoid using this time, allowing you to stay focused on what worked, rather than keeping what didn’t work in your chatbot’s memory for reference.
The fresh start is not only about “doing it right” with your chatbot this time, but also about resetting your creative flow.