Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
ghuntley
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
11 ms
·
181.
▲
LLMs are mirrors of operator skill
(ghuntley.com)
47 points
by
ghuntley
1y ago
|
90 comments
182.
▲
Vibe Coding Is Driving
(jasonpearson.dev)
2 points
by
ghuntley
1y ago
|
2 comments
183.
▲
Deliberate Intentional Practice
(ghuntley.com)
3 points
by
ghuntley
1y ago
|
0 comments
184.
▲
by
ghuntley
1y ago
ditto - https://ghuntley.com/disclosures
185.
▲
A Emoji Reverse Polish Notation Calculator Written in COBOL
(github.com)
4 points
by
ghuntley
1y ago
|
0 comments
186.
▲
From Luddites to AI: The Overton Window of Disruption
(ghuntley.com)
2 points
by
ghuntley
1y ago
|
0 comments
187.
▲
Someone got an LLM running on a Commodore 64 from 1982, and it runs as well
(xda-developers.com)
52 points
by
ghuntley
1y ago
|
17 comments
188.
▲
Xogot at Pax East
(blog.la-terminal.net)
1 points
by
ghuntley
1y ago
|
0 comments
189.
▲
by
ghuntley
1y ago
This is me - speak soon. https://www.linkedin.com/in/geoffreyhuntley
190.
▲
by
ghuntley
1y ago
the new Claude code “max plan” would last me all of [1] 5mins… I don’t get why people are excited about this. High powered tools aren’t cheap and aren’t for the consumer… [1] https://www.youtube.com/live/khr-cIc7zjc?si=
191.
▲
Elixir's Advantage in the Era of AI
(sylverstudios.dev)
4 points
by
ghuntley
1y ago
|
0 comments
192.
▲
When "Open" Isn't Open Enough: Rethinking Trust in Dev Tools
(blogs.eclipse.org)
1 points
by
ghuntley
1y ago
|
0 comments
193.
▲
Hacker 'NullBulge' pleads guilty to stealing Disney's Slack data
(bleepingcomputer.com)
2 points
by
ghuntley
1y ago
|
1 comments
194.
▲
Temu v0: Coding Agent
(github.com)
3 points
by
ghuntley
1y ago
|
0 comments
195.
▲
The secret recipe of powerful AI coding Agents
(madhukarkumar.medium.com)
13 points
by
ghuntley
1y ago
|
0 comments
196.
▲
LIVE: 4 Headless AI agents vibecoding Erlang/Elixir/Rust whilst I sleep [video]
(youtube.com)
2 points
by
ghuntley
1y ago
|
0 comments
197.
▲
Four agents vibecoding Erlang/pheonix whilst I sleep
(twitter.com)
1 points
by
ghuntley
1y ago
|
0 comments
198.
▲
Evaluating AI's Impact on Haskell Open Source Development
(well-typed.com)
6 points
by
ghuntley
1y ago
|
0 comments
199.
▲
The MCP Authorization Spec Is a Mess for Enterprise
(blog.christianposta.com)
1 points
by
ghuntley
1y ago
|
0 comments
200.
▲
Daytona Cloud: The Agent-Native Infrastructure
(daytona.io)
2 points
by
ghuntley
1y ago
|
0 comments
201.
▲
The Future of AI Belongs to Experienced Operators with Good Taste
(aaronstannard.com)
2 points
by
ghuntley
1y ago
|
1 comments
202.
▲
by
ghuntley
1y ago
> Every single business that has a need for SWEs will obligate their SWEs to use AI coding assistants by the end of 2026, if not by the end of 2025. Exactly why I authored https://ghuntley.com/ngmi - it's already ha
203.
▲
Anthropic Incident: Elevated errors on request to models
(status.anthropic.com)
4 points
by
ghuntley
1y ago
|
0 comments
204.
▲
by
ghuntley
1y ago
I'm going to get downvoted for this, but you can literally rebuild Kafka via AI right now in record time using the steps detailed at https://ghuntley.com/z80 . I'm currently building a full workload scheduler/
205.
▲
by
ghuntley
1y ago
you are correct. vscode (mit) is utterly evil - see https://news.ycombinator.com/item?id=43786380
206.
▲
by
ghuntley
1y ago
Thanks for the share. Yeah, building upon VSCode (MIT) is a stupid idea. Regarding OpenVSX, it was developed whilst I was at Gitpod and transferred to the Eclipse Foundation. It's been many years now, so my memory might be a little dat
207.
▲
Rizzler: Stop crying over Git merge conflicts and let AI handle the drama
(ghuntley.com)
1 points
by
ghuntley
1y ago
|
2 comments
208.
▲
Programming in the Age of Abundance
(guyren.me)
3 points
by
ghuntley
1y ago
|
0 comments
209.
▲
Autoregressive Queens of Failure
(ghuntley.com)
1 points
by
ghuntley
1y ago
|
0 comments
210.
▲
by
ghuntley
1y ago
See also https://isdotnetopen.com and https://ghuntley.com/fracture
More ›