swordsith

6 ms·

31.
by swordsith 5mo ago
Hopefully this means opeani wont exclusively distribute codex app through microsofts drm system
32.
by swordsith 6mo ago
Very obviously AI written, could tell by the end of the first sentence. Could've been a interesting read.. if you wrote it.
33.
by swordsith 6mo ago
Something about a scheduling/productivity app (one of the most common vibe-coded projects people make) being the subject of this is funny to me. I wonder how many tokens have been wasted making apps like this, let alone time.
34.
by swordsith 6mo ago
You're really completing bug bounties with found with AI? are companies honoring these?
35.
by swordsith 6mo ago
if michalesoft wants to take away our ability to sign drivers, they will find there is more than enough vulnerable easily exploited drivers we can use that are pre-signed online. Thank you micosawft!
36.
by swordsith 7mo ago
was a very interesting approach to web stack RE, I appreciated it. Good luck.
37.
by swordsith 7mo ago
"But it was still unclear how much we should trust this result because it was possible that at least some of those historical CVEs were already in Claude’s training data." I feel like they could know this if they truly wanted to.
38.
by swordsith 7mo ago
This model was not so fun to use for me, had it make a fancy landing page and sometimes it would forget about what i just asked it to do and affirm something it had done before was working. Just odd, needs too much hand-holding compared to
39.
by swordsith 7mo ago
I wish hah puzzling over the correct unwrap option is a pain
40.
by swordsith 7mo ago
bot
41.
by swordsith 7mo ago
This is my general experience with the claude app, I don't know what they're smoking over at anthropic but their ability to touch mobile arch inappropriately with AI is reaching critical levels.
42.
by swordsith 7mo ago
This is no different than a AI inference loop, just using a animal as a figurative code hamster in a wheel. The fact that the pre-prompt alone is this long in my opinion discredits any possibly interesting thing about this concept, So i wil
43.
by swordsith 7mo ago
I use a ML model i trained off of AI and human text, it flags mostly as Claude. The article even references claude.
44.
by swordsith 7mo ago
It looks like claude signed this at the end of the article for you 'ai, tools, opinion' lol
45.
by swordsith 8mo ago
Definitely generated, testing a ML model I'm training. it flags as claude mostly.
46.
by swordsith 8mo ago
Truth. I've had much easier time grappling with code bases I keep clean and compartmentalized with AI, over-stuffing context is one of the main killers of its quality.
47.
by swordsith 8mo ago
You should have told claude to summarize Claude.md instead of having it write this article, which by the sounds of it, it added to Claude.md.
48.
by swordsith 8mo ago
The cursor CLI was a great springboard for me to make my own frontend to prompt from, I would never use it though god, I implemented git, difs, and chat histories, so i can use it easily from my phone with tailscale. It picks up on my rules
49.
by swordsith 8mo ago
There is a stark difference between the behavior you can get out of a Chat interface LLM, and its API counterpart, and then there is another layer of prompt engineering to get around obvious censors. To think someone who plays with AI to me
50.
by swordsith 8mo ago
the buck on engineer status in my opinion stops when someone becomes a crypto scammer.
51.
by swordsith 8mo ago
Cursor does this well with its planning feature Imo.
52.
by swordsith 8mo ago
Not to dog on you but I've always had a feeling that password managers or any method of saving passwords are probably the worse net security vulnerability you can open yourself up to. Very silly concept.
53.
by swordsith 8mo ago
good take, I wish opus 4.6 wasn't so pricy its great for planning.
54.
by swordsith 8mo ago
Anyone not using version control or a IDE that will keep previous versions for a easy jump back is just being silly. If you're going to play with a kid who has a gun, wear your plates.
55.
by swordsith 8mo ago
I'm not sure if I follow, are the PRs legitimate and they are just being made to buff their resume, or are PRs malicious?
56.
by swordsith 8mo ago
Used ~1-3% of my cursor sub with gemini flash last night, made a personal image hosting front-end hosted with python to use from my phone with Tailscale. Took all of maybe the better part of a hour and a few prompts saying 'add this th
57.
by swordsith 8mo ago
Depends on the privacy practices of the people hosting the models, or are providing access to them. Most have an 'opt-out' of helping to train models.
58.
by swordsith 8mo ago
I've found mostly for context reasons its better to just have a grand overview of the systems and how they work together and feed that to the agent as context, it will use the additional files it touches to expand its understanding if
59.
by swordsith 8mo ago
If you've got a plan for the plan, what else could you possibly need!
60.
by swordsith 8mo ago
He would be! If they didn't get banned from using it in the OAI offices. He's so mad.

More ›