Micro Agent, a New AI Coding Agent, in 60 Seconds

preview_player
Показать описание

#javascript #typescript #react #figma
Рекомендации по теме
Комментарии
Автор

so its tdd for ai

btw this is what I mean when I say LLM's are low level components, they arent bad, we're just not building actual tools with it

AlexAegisOfficial
Автор

So, when the test is shit your code will be shit as well 😂

msschubi
Автор

Part of the motivation to code and learn to code is the feeling of solving a problem. this takes away the joy of coding - lol - but brilliant idea!

michaeltse
Автор

Writing good tests means you understand what you need. But our job is usually spent thinking what you need. Sure AI can perfectly define an email validator but not all problems are not well defined like that.

getrektnow
Автор

So now I have to sit there debugging/rewriting the tests instead of the code 👍

Aubstract
Автор

It's interesting to see the new Micro Agent AI coding agent! There are additional programs available that may provide even more functionality. #AIDevelopment #AI Coding

sirishkumar-mz
Автор

Cool idea, it would be good to have something like TDD agent that helps you iterate on tests and keeps on adding functionality.

cloud-native-corner
Автор

I was writing a research paper on one of my projects which does similar thing. U have an architect which breaks the problem statement into sub modules. Testwr weites tests for these modules and then the coder codes. This should remove the messups which can be caused usually with coder agents

agi_lab
Автор

nice. now we need a tool to test the test to make sure the test is correct

danielmontilla
Автор

I keep hearing that people get AI code that don't work. And i am honestly surprised, alot (not all) of the code i get from AI works. But yes there is some tweaking, but it sure as hell beats stack overflow.

martinolsson
Автор

Why are developers so determined to use the OpenAI api to write anti-intern / company lay off tools? Coding and writing tests is like the one aspect of being a swe that is fun and why most of us wanted to become swe.

emperorpingusmathchannel
Автор

Very interesting! But how's the quality of the output code? It's trivial to write code that can pass tests, but not so easy to write efficient, readable, maintainable, flexible, extensible, scalable, accessible etc. I often see AI falling short in this light, which is understandable given the typical quality of its training data. I'd also question the quality of the test passes, perhaps passing it through coverage and mutation analysis would be beneficial, although I suspect it could struggle if it gets stuck in local minimas in its solution space.

ShortFilmVD
Автор

i am creating ai tool that decreases code writing time and increases engineering time for dev, this is almost same approach i am using because at the end this is what i think is good solution

parshvsheth-rjvq
Автор

just tried it on my project. It installed fine, but on attempting to run it, it does not give me the startup blurb. I am running a windows machine. on terminating the application it gives me output `stopping` in red

sourmans
Автор

Does it code only JS or other languages as well?

marshallhiggins
Автор

Cool. We even more closer to AI replace all programmers and software engineers 👍

mndtr
Автор

How do you stop it if you made a mistake?

ZaryPhotography
Автор

Hey i am in serious problem after chatgpt i became a gpt based developer (meaning i just straight away hop on to gpt wherever i want to build our solve problem) i want to come out of it. What to do

aftabh
Автор

Before, ai was used as SDE, now its an SDET 😂 .

I liked you approach, but worst case it can take more time than normal .

But best case it improves the code quality . 😊

R_Y_Z_E_N
Автор

Tell me you never used ai to create tests without telling me you never used ai to create tests.

Iron_spider