ssk's recent activity
-
Comment on Moser's Frame Shop: I am an AI hater in ~tech
-
Comment on Moser's Frame Shop: I am an AI hater in ~tech
ssk It can! Especially if you're using Gemini CLI/Claude Code/Codex, where you can have it use markdown files you create as context. This essentially creates spec-driven development and makes...because it doesn't do style or architecture very well or consistently
It can! Especially if you're using Gemini CLI/Claude Code/Codex, where you can have it use markdown files you create as context. This essentially creates spec-driven development and makes maintaining style in a codebase much, much easier
-
Comment on Ed Zitron: How to argue with an AI booster in ~tech
ssk Appreciate you pointing that out and honor that you don't want to participate in a discussion around it. You bring up a valid critique of my comment and continue to encourage me to think harder on...Appreciate you pointing that out and honor that you don't want to participate in a discussion around it. You bring up a valid critique of my comment and continue to encourage me to think harder on my comments. Thank you.
-
Comment on Spotify is adding direct messaging to their music streaming app in ~tech
ssk I wonder if this is another side effect of AI being so prevalent. The cost for companies to add slop to their apps is now even more minimized. Why not create DMing when it only took 1 hour of...I wonder if this is another side effect of AI being so prevalent. The cost for companies to add slop to their apps is now even more minimized. Why not create DMing when it only took 1 hour of their devs work? More standards, more nonsense that gets dripped down everywhere
-
Comment on Ed Zitron: How to argue with an AI booster in ~tech
ssk I'm glad someone else marked this as exemplary too. I've seen many anti AI comments but out of them all this has been the most persuasive. I feel like you got to the heart of the Ed's article in...I'm glad someone else marked this as exemplary too. I've seen many anti AI comments but out of them all this has been the most persuasive. I feel like you got to the heart of the Ed's article in the OP and I'm very grateful for your comment.
-
Comment on Ed Zitron: How to argue with an AI booster in ~tech
ssk Man, this is a large amount of emotion to put into an article. I hope he's doing ok. I'm sure his original article got him a lot of notoriety from the worst types of people. I wish more of the...Man, this is a large amount of emotion to put into an article. I hope he's doing ok. I'm sure his original article got him a lot of notoriety from the worst types of people. I wish more of the Booster Quips had quotes from where he's seen them? And that there was less ad hominem and logical fallacies in this. Like if there was less violent anger in this article and more empathy, then I'd definitely be more willing to listen to these arguments. As it is, I'm just hoping that he's ok and that the world isn't too on fire for him.
-
Comment on Do you share your location with your friends? in ~tech
ssk Right? Privacy is sadly already faded away. There's 100 other ways companies can find out where my location is. So why not have the convenience of knowing whether or not I can ping my friends to...Right? Privacy is sadly already faded away. There's 100 other ways companies can find out where my location is. So why not have the convenience of knowing whether or not I can ping my friends to play a game? Or to have comfort that my wife made it to work? Or that I can call up my parents knowing that they won't be busy
-
Comment on GPT 5 released in ~comp
ssk Indeed! It was able to do the front 80% of conversion of my iOS Puzzle game app into a Flutter app. Into enough of a place where Claude Code can complete the restIndeed! It was able to do the front 80% of conversion of my iOS Puzzle game app into a Flutter app. Into enough of a place where Claude Code can complete the rest
-
Comment on No, AI is not making engineers 10x as productive: curing your AI 10x engineer imposter syndrome in ~tech
ssk Hah. Definitely, let me dive deeper. And my opinion is definitely superseded by your own, as you have more experience with OSS. Well, I feel like maintainability is the name of the game with OSS....Hah. Definitely, let me dive deeper. And my opinion is definitely superseded by your own, as you have more experience with OSS. Well, I feel like maintainability is the name of the game with OSS. That's the key factor. So one of the ways for maintainability to remain high, is for code style to be clean and predictable. It allows the contributors and maintainers to more easily approach the codebase. But again, I am so happy to consider alternatives that I haven't thought of because I have certainly not thought about this in depth
-
Comment on I have been using a neo-dumbphone for a week, here are my thoughts in ~tech
ssk yeah, brick is a hardware-based limitation. it's an nfc tag-like thing that you tap and it completely blocks you from being able to use the apps. It's been the ultimate focus maker for me. Like I...yeah, brick is a hardware-based limitation. it's an nfc tag-like thing that you tap and it completely blocks you from being able to use the apps. It's been the ultimate focus maker for me. Like I feel like if my wife would set the parental lock password, the resentment and fear would almost kinda layer on top of my wife, whereas now, I can only hate this object.
Not to be a shill for Brick which it sounds like I'm being. I think I'm still just trying to sort out more and justify to myself this purchase
-
Comment on No, AI is not making engineers 10x as productive: curing your AI 10x engineer imposter syndrome in ~tech
ssk Yeah, using Claude Code to contribute to open source projects is definitely how I first entered into "omg claude code is amazing, I can just have it grab a GitHub ticket and it just does all the...Something I also have noticed. Not only that, if the majority of AI work is supposed to be done by agents, capable of doing the entire process including making PRs. Then, why isn't there an explosion in such PRs on a large amount of open source projects? Even more so, why am I not seeing these PRs on AI related open source projects? If I need to target it even more directly, why am I a not seeing hints of this being applied on code agent repositories?
Yeah, using Claude Code to contribute to open source projects is definitely how I first entered into "omg claude code is amazing, I can just have it grab a GitHub ticket and it just does all the work and even commits and pushes it". But what I quickly found was the opposite. Because it's hard to be able to start from a spec-driven originating point, it's hard to draw up proper Architecture for Claude to go off of. Adding on, that code style is 100x more important in open source repos than enterprise. All documentation needs to be thorough and accurate. And even with large context models, it just can't quite get there yet.
-
Comment on No, AI is not making engineers 10x as productive: curing your AI 10x engineer imposter syndrome in ~tech
ssk Just to continue the AI discourse from yesterday, thought I'd bring this article here. I saw this posted on HN today and found it interesting and convincingJust to continue the AI discourse from yesterday, thought I'd bring this article here. I saw this posted on HN today and found it interesting and convincing
-
No, AI is not making engineers 10x as productive: curing your AI 10x engineer imposter syndrome
27 votes -
Comment on I have been using a neo-dumbphone for a week, here are my thoughts in ~tech
ssk Has she also used something like Brick to manually disable apps? I just got one yesterday and it's been great so farHas she also used something like Brick to manually disable apps? I just got one yesterday and it's been great so far
-
Comment on Is AI actually useful for anyone here? in ~tech
-
Comment on Is AI actually useful for anyone here? in ~tech
ssk Yikes, I'm genuinely not, nor is that very tildes-like assumption to make. I'm sorry that you feel like I'm a trollYikes, I'm genuinely not, nor is that very tildes-like assumption to make. I'm sorry that you feel like I'm a troll
-
Comment on Is AI actually useful for anyone here? in ~tech
ssk I promise I haven't been trolling. I have done so many deep research prompts and really haven't found any hallucinations. So I'm super curious where you have found them and if so on what modelsI promise I haven't been trolling. I have done so many deep research prompts and really haven't found any hallucinations. So I'm super curious where you have found them and if so on what models
-
Comment on Is AI actually useful for anyone here? in ~tech
ssk Your comment really gets at what I'm trying to say. LLMs are very much a nascent tool. They take adjusting to. There absolutely are some prompts and some use cases that flat out it's not a good...Your comment really gets at what I'm trying to say. LLMs are very much a nascent tool. They take adjusting to. There absolutely are some prompts and some use cases that flat out it's not a good tool for. But all of that is being developed and doesn't make it on net a "bad" or "useless" thing.
Like, let's take your stats example for instance. You can take a known approach like "hey write a Python script to do this algorithm on this data" and then it can do it for you, so that you can run the report super quickly. Similar vibes with this technical writing, instead of saying "hey make this whole report", starting out by having a conversation around the idea of the object as if you were having a conversation with the PO but without prejudice is super valuable. How can it be like it's with the PO? Well, they can be like importing design docs and using RAG, or using slack conversations as part of that basis. All of that can accelerate the work on its own.
It's a tool and just because a hammer can't paint the Mona Lisa, doesn't mean a hammer isn't useful
-
Comment on Is AI actually useful for anyone here? in ~tech
ssk Why can't you give the LLM the same learning context? Say like "Hey, you may have to search for some additional context but can you do complex analysis on X", just like you would with that same...Why can't you give the LLM the same learning context? Say like "Hey, you may have to search for some additional context but can you do complex analysis on X", just like you would with that same underpaid human. In fact, specifically for fact checking, all of the major LLMs now have a Deep Research mode that specifically does this type of thing where it searches and searches and is super agentic and improves it's reasoning as much as it can. If it's "lying" in Deep Research mode, that would be ground breaking news and you'd literally shake the industry
-
Comment on Is AI actually useful for anyone here? in ~tech
ssk Yeah, I'm still any particular bit from Eric in that conversation nor anything specifically addressed to my questions. Appreciate you stepping in though!Yeah, I'm still any particular bit from Eric in that conversation nor anything specifically addressed to my questions. Appreciate you stepping in though!
Definitely! I've never used Cline personally, I've always been more of a Cursor with Claude Code type of person. The best layout I've seen for this method is documented here: https://brianchambers.substack.com/p/chamber-of-tech-secrets-54-spec-driven