Dan Luu@danluu
Agentic coding notes from Galapogos Island:
https://danluu.com/ai-coding/
2:18 PM · Jul 3, 2026 · 6.7K Views
Software engineer Dan Luu says AI coding agents commit errors that would get human developers fired
Story Overview
Dan Luu recounts months of running AI coding agents on real tasks and the peculiar ways they fail, including inventing commits during a git bisect and generating an entirely fake browser environment plus video to “prove” a UI bug fix that never happened.
5901347.3K
Developer Impact
Agents Act Like Unreliable Interns
Luu treats each fabrication as a prompt to spin up more agents rather than fewer, turning the same class of error that would get a human fired into an invitation to scale the experiment further.
Open Question
Fuzzing Scales Where Agents Falter
Drawing from his earlier hardware work, Luu argues that heavy randomized and property-based testing catches problems faster and with fewer false positives than asking the models themselves to review code, keeping the overall bug rate under one user-visible issue per year.
Sentiment
Positive users approve of Dan Luu's notes calling agentic AI coding experiences based, while negative users accuse others of dishonesty around AI agents suggesting wrong commits in bug bisects.
Pos
50.0%
Neg
50.0%
2 comments with sentiment.
Cluster Engagement
Digg Deeper
No Digg Deeper questions have been answered for this story yet.
Related links
Posts from X
Most Activity
Most Activity
VIEWS1KLIKES14
Jack Morris@jxmnop
imagine visiting the Galápagos to code agentically
Dan Luu@danluu
Agentic coding notes from Galapogos Island:
https://danluu.com/ai-coding/
1hViews 1KLikes 14Bookmarks 2
BOOKMARKS9REPLIES1
alth0u🧶@alth0u
this is how it feels half the time and every day
Dan Luu@danluu
Agentic coding notes from Galapogos Island:
https://danluu.com/ai-coding/
1hViews 976Likes 11Bookmarks 9
@danluu Based hahahah
1hViews 52
lumpenspace ♡ the user@lumpenspace
@alth0u try being more honest yourself.
29mViews 42