This part kicks off a series where I dive into AI-assisted coding. In this first part I will dive into general lessons learned from 2 years of working together with AI to code.
Really aligned with this; I've been having a similar experience Sjoerd! You certainly cannot let it code like an agent, it does take a lot of iteration to find code that works. But I've found that it forces you to look at the code and actually learn coding. It's like the best teaching tool ever: teaching by doing by teaching by doing by teaching.
Absolutely, it's also incredible how much more accessible it has become to build full on functional projects with little to no actual coding experience. Those times where you come across something you've seen before and you can use that knowledge to get to the result you want are also some great!
I'm always working on a bunch of stuff! I'm currently working on semi-automated report generation for a client. Imagine a user being able to log in, record some audio, have it be transcribed, drag and drop a few extra relevant document and out rolls a word document that contains the desired report. Through the audio part the person still makes the calls, but all the copy and paste and filling in the gaps while also converting the language to the right style is handled by the LLM.
Really aligned with this; I've been having a similar experience Sjoerd! You certainly cannot let it code like an agent, it does take a lot of iteration to find code that works. But I've found that it forces you to look at the code and actually learn coding. It's like the best teaching tool ever: teaching by doing by teaching by doing by teaching.
Absolutely, it's also incredible how much more accessible it has become to build full on functional projects with little to no actual coding experience. Those times where you come across something you've seen before and you can use that knowledge to get to the result you want are also some great!
What are you building right now?
I'm always working on a bunch of stuff! I'm currently working on semi-automated report generation for a client. Imagine a user being able to log in, record some audio, have it be transcribed, drag and drop a few extra relevant document and out rolls a word document that contains the desired report. Through the audio part the person still makes the calls, but all the copy and paste and filling in the gaps while also converting the language to the right style is handled by the LLM.