IDE-native AI coding tools far outperform standalone chatbots
Thomas Domke · GitHub CEO Thomas Domke -- The One-Person, Billion-Dollar Startup · Jul 24, 2024 · at 7:06
Thomas Domke, CEO of GitHub, explains why developers benefit more from integrated coding assistants like GitHub Copilot than from copying code into generic web chatbots like Claude or ChatGPT.
0:00 / 0:44exact quote · 44.3s
▶ Watch the full episode on YouTube → 720p mp4 · rendered on demand · StarZero watermark
“The power of Copilot is that it lives, you know, in the work environment of the developer. So yeah, you can copy and paste everything that you see in front of you into a generic chatbot and have it give you an answer, but it's much more powerful to have the chatbot sit within your environment where it knows, you know, what files are open and knows what you wrote, you know, before that it can look at adjacent tabs. It can even look, you know, in in developer in the developer world at the output as a debug output is what we call that in the console and error messages and those kind of things. And so it has much more context available that helps it to answer the question, you know, that in the specific context of the project you're working on”
quote is from the automated transcript, cleaned for reading: filler sounds and stutters are removed, nothing is rephrased. names can be misheard (the analysis reads context, assessments check outside sources). how →
More from Thomas Domke
Large-scale software development with AI still requires expert human architects
“For small projects, you can probably you know, get there even without a lot of computer science, computer engineering knowledge for larger projects. I think the step missing is the architect, you know, the software engineering expert that knows which database …”
AI creates more work for developers, not less
“In fact, I'd say, you know, AI has created more work for developers because now somebody has to build all these AI systems, and we're not
At all at a point where you can just, you know, have an AI engineer, quote unquote do the job of a human, like that doesn'…”
AI will write 80% of all code by 2027
“I believe so. Yeah. I said actually two years ago at a conference that my prediction back then was 80% of code is going to be written by AI in five years. So I guess I have three years to go to, for that to become true.”
Global software developers will reach one billion by 2030
“And I think AI is going to accelerate that massively. And, you know, one billion developers by 2030 or so is a little bit under 10% of the population, depending on where the world's population is going. That's actually a low number if you think about it, becau…”
AI will not replace code, but push it to a lower abstraction layer
“And we still have code. Look, look, the AI and the you know, Copilot is not going to replace the code. The code is just lower in the abstraction level in the same way that, you know, your chip in your computer still has an instruction set, you know, we used to…”
A one-person, billion-dollar company is feasible
“I think so. I think that, you know, the, you know, I've seen a bunch of examples where small companies like Instagram comes to mind, you know, that, that started really small and by the time they got acquired, they were still very small. WhatsApp is a similar …”