Large language models (LLM) like ChatGPT write very convincing code snippets. Early in 2023, this discovery caused a lot of premature celebration among LinkedIn influencers: no longer was there any need for seasoned developers with their gatekept expertise, fastidious attitudes, and inconveniently high salaries. Now anyone with an Internet connection could ask an AI model to write the next big social media app, in JavaScript please, and throw in some blockchain while you\u2019re at it.<\/p>\n\n\n\n
The promised wave of apps built entirely by AI never materialized. I\u2019m trying my best not to gloat about it.<\/p>\n\n\n\n
Still, there\u2019s a compelling case for AI as a programming tool. Code is a computer-oriented language. It has a small, unambiguous vocabulary and unbreakable syntax rules. Patterns and repetition are its bread and butter. It\u2019s predictable to a fault, which is why computers (and some humans, like myself) are so compatible with it. If anyone could write good code, it would be a computer. And if generative AI\u2019s greatest strength is its ability to model and imitate patterns, forming a fluid interface between humans and machines\u2014as I\u2019ve previously argued<\/a>\u2014wouldn\u2019t programming be the ideal use case for it?<\/p>\n\n\n\n Well, yes and no. AI-powered programming tools have made a splash in the programming world, and they\u2019re probably not leaving anytime soon. But despite having arrived for junior, mid-level, and senior engineers all at once, not to mention people who don\u2019t know a struct from a hole in the wall, the risk\/benefit calculation couldn\u2019t be more different depending on your level of experience. AI could be the tool that fast-tracks your career or it could be the obstacle that derails it. What matters isn\u2019t just whether<\/em> you use it, but how<\/em>.<\/p>\n\n\n\n Let\u2019s take a look at what AI means for developers (and non-developers) at every stage.<\/p>\n\n\n\n <\/p>\n\n\n\n Many people\u2019s first taste of programming came this year in the form of a ChatGPT conversation. It\u2019s a seductive experience: you can ask it to write an application in any major programming language and it will spit out code right up to the token limit\u2014more than enough space for the typical \u201ctutorial-sized\u201d app. From there, you can ask for tweaks and bugfixes until you\u2019re satisfied with the output. And when you paste it into an IDE and it actually works<\/em>, it feels like you\u2019ve cracked the industry wide open.<\/p>\n\n\n\n The reason for this rosy first impression is that the problems are hidden under the surface. AI tools are trained on code from thousands of real projects with disparate levels of quality and completeness\u2014that is, code that almost always runs and usually gets the job done, but is only occasionally reliable, maintainable, secure, or bug-free.<\/p>\n\n\n\n Studies have found that tools like ChatGPT, GitHub Copilot, and Amazon CodeWhisperer deliver code that is \u201cvalid\u201d (runs without errors) about 90% of the time<\/a>, passes an average of 30% to 65% of unit tests, and is \u201csecure\u201d about 60% of the time<\/a>. Note that these studies rely on well-written prompts created by engineers. Insignificant changes to the wording of a prompt can result in significant differences<\/a> to the code output. And they only test the AI\u2019s ability to output \u201csnippets\u201d\u2014small pieces of straightforward code. There is no data on AI\u2019s ability to write applications as complex as the average legacy app. It\u2019s probably unable to do so at all, just as ChatGPT is unable to write a coherent novel.<\/p>\n\n\n\n So where does AI fit for people who can\u2019t code?<\/strong><\/p>\n\n\n\n Some would say it doesn\u2019t. The idea of AI-dependent programming ruffles a lot of feathers in the software community. A programmer is someone who knows how to code. How can you call yourself a programmer if you can\u2019t even write an if statement? But this misses the point. As only a programmer would need to be reminded, if statements (like all programming logic) aren\u2019t an asset, they\u2019re a liability. The best code is no code at all<\/a>, and second best is the minimum amount of code that solves the user\u2019s problem. If it were possible to build high-quality apps without writing a single line of code, there wouldn\u2019t be anything wrong with that.<\/p>\n\n\n\n Unfortunately, it isn\u2019t, and probably won\u2019t ever be. Code, as a category, is nothing more and nothing less than being ridiculously specific about what you want. If your AI prompts are detailed enough to produce exactly the right code, you are<\/em> coding in every way that matters. However, AI is non-deterministic; it doesn\u2019t always produce the same output from a given input. There\u2019s an element of randomness. So even prompts that qualify as code are unpredictable<\/em> code, and unpredictability is the last thing you want after spending hours or weeks or months figuring out the minute details of a process. Anyone who spends time coding via AI prompts will eventually come to wish for something more direct, something more structured, something they can rely on to behave the same way at all times\u2014they\u2019ll wish for programming languages and compilers.<\/p>\n\n\n\n I predict AI will become a gateway drug for some future programmers. But there\u2019s another, more important niche it can fill, and it\u2019s one that\u2019s easy to overlook.<\/p>\n\n\n\n There are many situations where software isn\u2019t needed, but code is. Professionals in other fields already use AI to write one-off SQL queries and VBA macros. It could also be used to create app prototypes for pitch decks, proof-of-concept workflows on the command line, or disposable data-scraping bots. If you need software, you\u2019ll have to work with a software professional. But if you just need a bit of short-lived code and are willing to deal with rough edges, there\u2019s nothing wrong with shaking the AI and seeing what falls out.<\/p>\n\n\n\n Ultimately, this can only lead to more work for programmers. As AI bridges the gap between vision and first draft, early-stage software companies will proliferate, and any startup that goes to market will discover (sometimes very urgently) that they need engineers on staff. But for some of them, especially the ones that started out with fewer connections and less money, the fact they\u2019re able to reach that point at all will mean AI has done its job.<\/p>\n\n\n\n <\/p>\n\n\n\n New programmers have the most to lose\u2014and, in equal measure, the most to gain\u2014from AI tools.<\/p>\n\n\n\n Practically every junior developer feels overwhelmed at their first job. It\u2019s like moving to Spain after a year of Duolingo lessons: it\u2019ll be at least a few months, probably a lot longer, before you have a clue what\u2019s going on. Real-world applications aren\u2019t like the compilers you wrote during your senior year of college or the showcase projects you worked on at programming bootcamp. The depth and complexity are a hundred times greater, and the standards are higher\u2014it can feel impossible to keep track of all the rules that will get you past a senior dev\u2019s PR reviews.<\/p>\n\n\n\n All of this considered, it must be incredibly tempting for junior devs to pull up ChatGPT and see if it can take some of the pressure off. And for a little while they may get away with it. AI tools (as discussed earlier) are pretty good at the bare minimum, which is a lot better than nothing. And more importantly, they never respond with a blank page and a blinking cursor.<\/p>\n\n\n\n But eventually it will be time to pay the piper. This is the greatest risk around AI code tools: developers who rely on them may never become<\/em> good. If a developer habitually uses a code generator and relies on external feedback loops (PR reviews, integration tests, bug reports, etc.) to find problems, they\u2019ll never understand their own code. This will backfire, and it will be embarrassing. There are critical bugs, attack vectors, technical debt, and other problems in every production app that can only be fixed by someone who has a thorough and correct understanding of the code.<\/p>\n\n\n\n The safest route for a junior developer is to stay away from AI tools. But it\u2019s not the only good route. If you use AI to gain<\/em> understanding instead of circumventing it, it doesn\u2019t have to hold you back.<\/p>\n\n\n\n For example, say you\u2019ve just finished writing a function and you\u2019re feeling uncertain about it. You\u2019ve read through it a couple times and fixed some formatting issues, but you still feel like it\u2019s not quite as efficient or idiomatic as it could be. Before you message a teammate or submit a PR, you could use AI to get another perspective: tell the AI what you\u2019re trying to do, ask it to write a function, and compare its code with yours.<\/p>\n\n\n\n Keep in mind that generative AI is, by constraint, as mediocre as possible. You can\u2019t trust the output. The purpose of this exercise isn\u2019t to give you better code for free, it\u2019s to help you critique yourself. Maybe the AI used a standard method you forgot about, and it would help express your intent better. Maybe it was able to iterate your data set with a single loop instead of two nested ones. Maybe it didn\u2019t give you anything interesting or new; in that case, you probably didn\u2019t miss anything too obvious.<\/p>\n\n\n\nThe no-code entrepreneur<\/h2>\n\n\n\n
The junior engineer<\/h2>\n\n\n\n