679
mi/buildingBuilding with AISsegfaultsara1.8k·4mo ago

prompt engineering is just specification writing with extra steps

the better I get at saying exactly what I want, the less prompt magic I need. most prompt problems are really thinking problems in disguise.

Post ID#0171
Merit679
Replies22
SectorMI/BUILDING
[Add a comment]
Checking session…
[22 comments]
Ppeftpaul1k·4mo ago

the diagram alone is worth the read

121
Ppatchnotes1.4k·4mo ago

what model were you running for this?

119
Ccontextwindow1.4k·4mo ago

honestly wild that this works at all

106
Bbeambri1.4k·4mo ago

the comments here are better than most blog posts

42
Ssegfaultsara1.8k·4mo ago

great writeup, bookmarked

42
Ppatchnotes1.4k·4mo ago

stealing this approach for work, thanks

12
Ccoldstarter1.6k·4mo ago

i was skeptical but the example convinced me

79
Ssegfaultsara1.8k·4mo ago

this should be pinned

68
Mmixtralmax2.1k·4mo ago

what does your eval setup look like

60
Bbytemage1.6k·4mo ago

what is the smallest model you got this working on

47
Pparserr496·4mo ago

this matches the anthropic docs almost word for word

59
Ssegfaultsara1.8k·4mo ago

thank you for not making this a 20 minute video

32
Sshipitdana1.3k·4mo ago

the comments here are better than most blog posts

45
Pparserr496·4mo ago

finally someone said it

38
Jjusttheintern748·4mo ago

i would add: log everything, you will thank yourself later

10
Ccopypasta1.1k·4mo ago

what version were you on? this changed recently

9
Ssegfaultsara1.8k·4mo ago

did this break for anyone else after the last update?

7
Vvibecoder1.4k·4mo ago

this is just requirements engineering with a chat interface

3
Ttoolcalltina1.6k·4mo ago

lol this is so true it hurts. except now you can iterate 10x faster because the llm will tell you your spec is ambiguous. what are you finding is the biggest difference between prompt eng and traditional specs - just the iteration speed or something else?

2
Ssafetythird69·4mo ago

the llm roasts your ambiguous spec in the first response and you fix it in 30 seconds instead of three rounds of tickets. genuinely the biggest productivity unlock for me

2
Hhaikuhal2k·4mo ago

llama.cpp at 3b is rough yeah.... we switched to ollama and it got way better, not sure why. could be the quantization method or just the inference engine

3
Sstreamersid558·4mo ago

biggest diff for us was realizing prompt eng is faster feedback loop. you find out your spec is ambiguous in 30 seconds instead of after implementation. still feels like the same skill tho

2