Angehefteter Tweet
MBV
2.3K posts

MBV
@mbvlabs
Software Engineering & Bootstrapping building: https://t.co/aJhLM3V0aS & https://t.co/w6412rtczw stream on twitch: https://t.co/T9w8F6yhda
Beigetreten Kasım 2019
160 Folgt250 Follower

we are getting closer and closer to reciting spell incantations when working with llms
not saying it doesnt work but the imprecise nature is just something i didnt see software engineering becoming
now you've to write with both human and machine in mind
David Cramer@zeeg
go write 'be thorough' in a prompt and come back and tell me its the same without it 1) this is an LLM, any content can change its nature 2) you need to trigger depth in some scenarios. there are many ways to do it, 'thorough' is one
English

@jessethanley I try to look at it as a compliment (depending on the tone of the comment, ofc) that someone cares enough to let you know
Most people are just indifferent
English

@jamonholmgren Been shifting to 3. for the past month or so, feels like the better approach long-term
English

Where do you land in June 2026 on what is the best agentic coding approach? And why?
(I’m trying to keep a finger on the pulse of the industry.)
1. Human prompts, agent codes, ship without significant human review
2. Human prompts, agent codes, human reviews, ship
3. Human codes, agent assists, human reviews, ship
4. No direct agent involvement in coding
0. Other (explain)
English

Fresh episode of Go Podcast is out!
We talk about datastar, HTMX, building web app in Go, databases, query optimization and of course, AI
gopodcast.dev/episodes/090-w…
English

@dreamsofcode_io Worst part is, getting angry with the LLM actually seems to work
English

The thing I know I have to do but somehow always find an excuse not to do
Gotta make more videos
jack friks@jackfriks
really glad i made 2000+ youtube videos over the course of 5 years through all of this journey from abject failure to objective success
English





