Justinas Kairys’ Post

Over-reliance on AI can lead to a lack of understanding of the underlying code and make it difficult to troubleshoot and debug. Additionally, understanding the basics helps to develop problem-solving skills, critical thinking, and creativity, which are essential for growth and success in the field. #AI #codeAI

The Day Has Finally Come: Junior Devs Only Know ChatGPT! We knew this day was coming. They warned us. There was nothing we could do. Junior developers are joining the job market and they can't code! They can only speak to LLMs! ... Or is that really true? Let's jump over to this Reddit thread and see how a software engineer is perceiving this as junior developers join their team! 👇Check the comments for the video link!👇

This is spot on and exactly what I’ve been saying as well for a little while now. These latest models appearing are changing the game. I would just apply this concept even further. Not to just developers but really any domain. But particularly data professionals who are perfectly skilled to evolve and go up the complexity scale to new digital skills, like ai application development.

After using a variety of these AI tools, it feels like the skills beyond purely knowing syntax or a particular framework are becoming more valuable. Whereas coding in the past felt like starting from a (somewhat) blank canvas, now it feels more like sculpting from marble. The ability to discern and think critically have always been important components of the craft, and this has never been more true than now. However, claims that a single engineer can now do the work of 20 are an exaggeration. You can use LLMs to generate loads of spaghetti code and give the illusion of increased output, but that in itself is meaningless. As it stands now, no model can "plan, generate different chains of thought, perform MCTS on them and backtrack" faster than my brain can. What does this mean for the job market? Probably challenging things in the short term, given that we are at peak hype. Perhaps at some point, someone creates a superintelligence that gives us true cause for worry... but as Noam Brown at OpenAI said: "You don't get superhuman performance by doing better imitation learning on human data." And that's all that seems to be happening lately. #machinelearning #ml #softwareengineering

“Ever struggled to explain what you do to someone outside of tech? I asked ChatGPT to explain backend engineering as if I were a 10-year-old, and the result was too good not to share: ‘Imagine building a huge LEGO city. The front of the city is all the cool stuff people can see and play with—shiny cars, buildings, and LEGO people walking around. That’s the frontend—what people interact with. But behind the scenes, you need invisible “robot workers” to move stuff around, pipes to bring water to houses, and cables to power everything. That’s the backend. It makes sure everything works. In computers, backend engineering is like building those invisible systems. It’s the code that stores and sends information—like where game saves go, how YouTube finds your videos, or how messages go to the right person in a chat.’ This explanation not only simplifies backend engineering but also reminds me how powerful storytelling is, no matter the audience. Sometimes, breaking down complexity is the real skill! How would you explain your job to a 10-year-old? P.S. Turns out explaining tech like this is way more fun than I expected. Give it a try—you might surprise yourself!”

The Evolution of Junior Developer Life: SE 1.0 (001.0) Junior Dev: - 8am: Boot terminal, question career choice - 10am: Stare at cryptic legacy code - 2pm: Break production trying to fix a typo - 4pm: Hide from senior dev after breaking prod - 6pm: Leave wondering if programming is for you SE 2.0 (010.0) Junior Dev: - 8am: Update 12 npm packages - 10am: Google "how to center div" for 100th time 🤦♂️ - 2pm: Try to understand why microservices - 4pm: Accidentally push to the main branch 😱 - 6pm: Wonder which framework to learn next SE 2.5 (010.1) Junior Dev Today: - 8am: Ask ChatGPT to explain yesterday's code - 10am: Debug AI-generated code you don't understand - 2pm: Wonder if learning to code was necessary - 4pm: Prompt engineering > actual engineering - 6pm: Contemplate if AI will replace you before you become senior The only constant? Imposter syndrome 😅 They say that in SE 3.0 we'll need an "AI psychology" degree to debug AI's code... 🤖 #JuniorDev #Programming #TechHumor #CodingLife #AI

Let’s clarify a few myths about programming that often mislead aspiring developers: 1) "AI will help you become a programmer." Not exactly. If a machine is doing the coding for you, then you’re not a programmer you’re merely a middle person adding unnecessary cost to a business. This makes your role redundant. A true programmer understands and builds systems, not just relies on AI. 2) "Drag-and-drop tools and connecting components make you a programmer." Wrong. If you don’t understand the underlying architecture, you’re not a programmer. Google and Facebook weren’t built by drag-and-drop developers. They were built by programmers who could design frameworks, build scalable systems, and create tools like low-code platforms for non-programmers to use. Pay attention to the term non-programmers. 3) "You can learn programming using ChatGPT." Yes, you can. But it requires effort. ChatGPT can assist you, but you’ll never build scalable systems if you don’t understand what you’re doing. Learning the fundamentals and applying them rigorously is essential. 4) "Having a senior software engineer title makes you a senior developer." Not necessarily. Titles don’t define capability. I’ve seen people with fancy titles, like Chief Engineer, struggle to build scalable systems. Being a senior developer is about skill, experience, and the ability to deliver, not just a title. 5) "Programming is all about writing code." This is the biggest myth. Programmers aren’t just code monkeys. If you call yourself a software engineer, you have a bigger responsibility. Understanding requirements, designing scalable architecture, reviewing peers code, and mentoring others are critical skills for a true engineer. 6) "Programmers are not architects." This notion is completely misguided, and I blame the rise of SaaS businesses for diluting the term. Architecture and programming go hand in hand. Look at companies like Google and Facebook architects there have a deep understanding of technology stacks and a curiosity to ask why, not just how. Many are pretty hands on engineers as well. 7) "Programmers are not testers." This myth needs to go. Testing isn’t just the tester’s responsibility. Programmers must ensure their code is functional, clean, and scalable before handing it over. Testers aren’t there to clean up a programmer’s mess.

A strong community enhances a programmer's journey, offering invaluable benefits that technology alone cannot match. Here's why being part of a programmer's community remains essential in the age of AI and ChatGPT: 👉 **Human Insight:** Tackling real-world problems often requires the nuanced understanding that only human experience can provide. 👉 **Collaboration:** Communities foster teamwork, mentorship, and idea exchange, driving innovation. 👉 **Beyond Code:** Discussions on design, architecture, and user-centric decisions complement AI's capabilities. 👉 **Networking:** Building connections within communities opens doors to career opportunities and professional growth. 👉 **Emotional Support:** Communities offer motivation, support, and a sense of belonging, combating isolation. 👉 **Ethics and Best Practices:** Collective input helps set ethical guidelines and industry standards. Let's compare two key communities: ⏩ **GitHub:** Project-specific, repository-centered discussions for collaboration and community engagement. ⏩ **Stack Overflow:** General programming Q&A platform for solving coding problems with concise answers.