dnw

9 ms·

91.
by dnw 8mo ago
Type of conversation would be interesting? (e.g. planning, discovery, banter, etc.)
92.
by dnw 8mo ago
Yes, they can. We have gotten better at grounding LLMs to specific sources and providing accurate citations. Those go some distance in establishing trust. There is trust and then there is accountability. At the end of the day, a business&#x
93.
by dnw 8mo ago
It is a little more than semantic search. Their value prop is curation of trusted medical sources and network effects--selling directly to doctors. I believe frontier labs have no option but to go into verticals (because models are getting
94.
Welcome to the Room: A lesson in leadership by Satya Nadella(jsnover.com)
28 points by dnw 8mo ago|57 comments
95.
Claude Agent Skills: A First Principles Deep Dive (2025)(leehanchung.github.io)
3 points by dnw 8mo ago|0 comments
96.
by dnw 8mo ago
I have had this new car for 5 months. I haven't learned to turn on the headlights yet. It just turns itself on and adjusts the beams. Every now and then I think about where that switch might be but never get to it. I should probably kn
97.
by dnw 8mo ago
"We submitted detailed technical reports through their coordinated security reporting process, including complete reproduction steps, root cause analysis, and concrete patch proposals. In each case, our proposed fixes either informed o
98.
A macOS app that blurs your screen when you slouch(github.com)
692 points by dnw 9mo ago|223 comments
99.
Claude Code TeamateTool (binary analysis)(gist.github.com)
2 points by dnw 9mo ago|1 comments
100.
GPTZero finds 100 new hallucinations in NeurIPS 2025 accepted papers(gptzero.me)
8 points by dnw 9mo ago|2 comments
101.
Programming, Evolved: Lessons and Observations(github.com)
49 points by dnw 9mo ago|24 comments
102.
by dnw 9mo ago
Besides their music, their mail-order ticketing system and the resulting fan envelope art are quite amazing: [Apologies for the scary link] https://www.gdao.org/fan-art?filters[match]=all&filters[quer...
103.
by dnw 9mo ago
I'd note a couple of things: Not to nitpick but if we are going to discuss the impact of AI, then I'd argue "AI commoditizes anything you can specify." is not broad enough. My intuition is "AI commoditizes anything
104.
by dnw 9mo ago
I really want to see if someone can prompt out a more elegant proof of Fermat's Last Theoremthan, compared to that of Wiles's proof.
105.
by dnw 9mo ago
When I read the title I thought you made a challenge Claude couldn't solve but that's not what you are doing. You are taking a pragmatic approach to the world we live in. I like it. - It would be good to put what you are planning
106.
by dnw 9mo ago
That is a cool tool. Also one can set "cleanupPeriodDays": in ~/.claude/settings.json to extend cleanup. There is so much information these tools keep around we could use. I came across this one the other day: https:&#x
107.
by dnw 9mo ago
Not surprised (#5): https://news.ycombinator.com/item?id=46395714#46425529
108.
by dnw 9mo ago
Fair question. It is "imperative" for two reasons. The first, despite having rough edges now, I find these tools be actually useful so they are here to stay. The second, I think most developers will use them and make them part of
109.
by dnw 9mo ago
The author should try Google Local Services Ads instead of Google Ads. I think Google cannibalizes Google Ads with LSA.
110.
by dnw 9mo ago
I have been using Copilot, Cursor, then CC for a little more than a year now. I have written code with teams using these tools and I am writing mostly for myself now. My observations have been the following: 1) These tools obviously improve
111.
by dnw 9mo ago
I have seen Claude disable its sandbox. Here is the most recent example from a couple of weeks ago while debugging Rust: "The panic is due to sandbox restrictions, not code errors. Let me try again with the sandbox disabled:" I ha
112.
On the success of ‘natural language programming’(brooker.co.za)
1 points by dnw 10mo ago|1 comments
113.
by dnw 10mo ago
"I believe that specification is the future of programming" says Marc Brooker, an influential DE at Amazon as he makes the case for spec driven development in the blog post.
114.
by dnw 10mo ago
If you are on macOS it is not a bad idea to use sandbox-exec to wrap your claude or other coding agents around. All the agents already use sandbox-exec, however they can disable the sandbox. Agents execute a lot of untrusted coded in the fo
115.
by dnw 10mo ago
They all partner with Github to detect leaked credentials. In order to have API keys I need to have an account with each service with a valid email. So all three of them had the same information and channels available to reach me. It wouldn
116.
by dnw 10mo ago
Claude was looking up env-vars during the coding session which ended up in ~/.claude/projects/ log. I wanted to make the [construction] logs public with the code. Didn't think that was a leak vector.
117.
by dnw 10mo ago
- Download all the source code and look for vulnerabilities at their leisure. - Depending on whether they use GH for deployments they can also introduce features to production that can help them
118.
by dnw 10mo ago
Last week I accidentally exposed my OpenAI, Anthropic, and Gemini keys. They somehow ended up in Claude Code logs(!) Within seconds I got an email from Anthropic and they have already disabled my keys. Neither OpenAI nor Google alerted me i
119.
by dnw 10mo ago
I was thinking this is going to happen because last night I got an email about them fixing how they collect sales taxes. Having been part of a couple of IPO/acquisitions, I thought to myself: "Nobody cares about sales taxes until
120.
by dnw 11mo ago
I too am curious. My daily driver has been Claude Code CLI since April. I just started using Codex CLI and there are lot of gaps--the most annoying being permissions don't seem to stick. I am so used to plan mode in Claude Code CLI and

More ›