Paul Kim (@paulbkimdev) profile picture

All opinions are my own.

Joined July 2022
Photo by Paul Kim (@paulbkimdev)Photo by Paul Kim (@paulbkimdev)Photo by Paul Kim (@paulbkimdev)Photo by Paul Kim (@paulbkimdev)Photo by Paul Kim (@paulbkimdev)Photo by Paul Kim (@paulbkimdev)Photo by Paul Kim (@paulbkimdev)Photo by Paul Kim (@paulbkimdev)Photo by Paul Kim (@paulbkimdev)Photo by Paul Kim (@paulbkimdev)
Sol 6.2 is fantastic model for real work.
1
1
11
Just got my colleagues who are game developers on @t3dotcodes . Curious to see what they think about it. If they like it T3GameDev would be my next project.
84
I don't get it. What's the point of a desktop app if it doesn't support things that actually need a GUI, like browser rendering, real-time app simulation, or at least running VMs to test apps across different environments? Herdr is great as a terminal multiplexer, but most people who use multiplexers already rely on Vim and terminal-based PR tools anyway. I just don't see the value in a half-baked desktop experience.
Claude Code users, if you didn’t know, Herdr now has its 𝗼𝘄𝗻 𝗱𝗲𝘀𝗸𝘁𝗼𝗽 𝗮𝗽𝗽. Install it, and Opus 5.5 can coordinate 𝘂𝗽 𝘁𝗼 𝟭𝟬 𝘄𝗼𝗿𝗸𝗲𝗿𝘀 through Herdr: Claude Code writes the code, nitpick-obsessed Codex Astra reviews it, and the app comes with a perfect dashboard built in. Hand off the heavy implementation work to Claude Code, including polishing your pages’ layout, interactions, and visual details. Set it up once. Here’s how 👇 1. Open Claude Code and install Herdr and its Skill Select Opus 5.5 and ask it to run these two commands (Mac): curl -fsSL releases.herdr.dev/app/insta… | sh hc --skill Once installed, have Opus 5.5 confirm that hc works. 𝗬𝗼𝘂 𝗸𝗲𝗲𝗽 𝘄𝗼𝗿𝗸𝗶𝗻𝗴 𝗶𝗻 𝗖𝗹𝗮𝘂𝗱𝗲 𝗖𝗼𝗱𝗲. Herdr manages the workers in the background, so you don’t have to switch interfaces. 2. Have Opus 5.5 finish the setup Paste this prompt: “Use the official Herdr docs and the instructions from hc --skill and herdr-app --skill to create a coordination Skill called herdr-desktop. Keep the main conversation in this Claude Code session and use hc to manage background workers. Don’t require me to operate inside a Herdr terminal. Start separately named background workers for this task, explicitly setting the working directory. Check whether Claude Code CLI and Codex CLI are installed. If either is missing, install it. If authentication is needed, ask me to sign in. After authentication, start one Codex worker using Astra, with reasoning effort explicitly set to xhigh. Give it a read-only task: inspect the project and explain its main entry points. Run Codex with --no-daemon and allow network access in its sandbox, or it can’t reach hc. The first time it starts in a new folder, it asks whether to trust that folder. If it gets stuck there, tell me. Wait for it to finish and read back its actual response to confirm that task dispatch and result retrieval work. If the model or configuration is unavailable, tell me. Don’t silently substitute another. When the Herdr app is open, put the workers in a space so I can see them in the sidebar. Save the verified startup, dispatch, wait, and result-reading procedures in the Skill. Don’t modify project code during setup.” 3. Sign in to your Codex account Opus 5.5 will check Codex’s login status. If needed, follow the prompts to authenticate with your ChatGPT account. After that, you keep assigning tasks in Claude Code, while Codex runs in the background using your ChatGPT account’s allowance. 4. Tell Opus 5.5 how to delegate Paste this prompt: “From now on, use this workflow: You are the coordinator. → Handle project understanding, planning, task decomposition, integration, and final review. Make the best decision that will elevate all three: the user experience (UX), developer experience (DX), and agent experience (AX) always, without breaking anything. → When a task needs substantial implementation, debugging, test fixes, or multi-file changes, delegate it to Claude Code through herdr-desktop. For frontend work, have it refine the layout, visual details, and actual interactions. Give it this rule: Do not design for rare or future cases nobody asked about. Use Opus 5.5 for Claude Code workers and Astra for Codex workers, both with reasoning effort explicitly set to xhigh. Briefly explain the division of work, then start independent tasks in parallel. Choose concurrency based on the task, machine load, and available allowance, up to 10 workers. You don’t need to use all 10. Give each task a clear objective, working directory, necessary context, deliverable, and completion criteria. Use separate worktrees for parallel code changes. You own the integration. Send follow-up feedback to the original worker whenever possible and continue its existing conversation. Don’t start over every round. After a worker finishes, inspect the actual changes yourself and run the necessary checks after integration. For page changes, open the browser and check them. After code changes, hand the review to a Codex worker that didn’t work on the implementation, with this rule: Rate each bug from 1 to 10 on how often it would happen and how bad it would be. If the two scores multiplied are below 5, just file it as an issue and leave the code alone. Don’t treat a task as finished just because another agent says it is. At the end, show me the result, what was verified, and any unresolved issues. Release task resources that are no longer needed while preserving the work and records needed to continue.” A few tips: 1. Save the delegation instructions from step 4 in herdr-desktop too, so you can invoke the Skill at the start of future tasks. 2. Pair the Skill with a clear end goal. For example: “Finish this pricing page, keep it consistent with the existing site, and check the mobile layout and plan switching.” 3. Each account still has its own usage limits. Adjust concurrency to the task and remaining allowance. 4. For long tasks, keep a record of progress, key decisions, and next steps. When switching conversations, check the actual state before continuing. It works the other way around, too: Codex coordinates, and Claude Code executes. Swap the roles and their corresponding settings.
148
I asked ChatGPT to interpret OpenAI's latest video, "There are many futures. Which do you want to build?", through the lens of the Bogdanov–Takens bifurcation spotted by some viewers. Its interpretation is quite fascinating.
There are many futures. Which do you want to build?
4
58
So perhaps "There are many futures" isn't just a philosophical statement. What if OpenAI is hinting that we're approaching a bifurcation point in AI development itself?
3
Imagine AI designing better AI, which in turn becomes better at designing its successor. At some point, this feedback loop could change the rate and nature of technological progress itself. The breakthrough wouldn't just be a smarter model. It would be a fundamentally different mechanism of progress.
4
Now imagine those two parameters represent: AI's capability to conduct AI research. The degree of autonomy in that research process. As both improve, we could approach a threshold where AI development no longer operates under the same dynamics.
1
2
In dynamical systems, a Bogdanov–Takens bifurcation describes a critical point where small changes in two parameters can fundamentally alter a system's behavior. The important part isn't simply acceleration. It's that the system's underlying dynamics can change entirely.
26
Well, well, well. Grok 4.8
Replying to @techdevnotes
I know it’s taken longer than expected, but Grok 4.8 will not be a disappointment. We’re not optimizing for benchmarks and I really don’t care how Grok does on those, but it will be incredibly useful.
30
Grokbot should really target Korea. Koreans are so fast to adapt these kind of technology and they just never move between services. The first kind to come into this market will literally have monopoly. @poteto
23
I want open-weight models to thrive, but man the community is so out of touch. they talk like self-hosting is a no-brainer when the economics don’t work with what hardware costs. and “only a few months behind” means nothing when the people you’re competing with are on the frontier models
1
26
to be fair they’re great for stuff that doesn’t need much intelligence. classifiers, embeddings, search, cheap image/video gen for non-commercial stuff. just stop pitching them as a frontier replacemen
18
`Agents per Capita` will be a thing. Mark my words.
14
ChatGPT Codex is great value as long as you can put on top of Claude 20x max sub, Grokbot Sub. They are good for "reivewing" stuff.
1
1
63
I can't agree more.
Codex Pro 200刀我感觉非常不经用,体感下来只有 Claude Max 200刀账号1/4不到,前者使用 GPT-6 Astra Medium,后者使用 opus 5.5 high。 有多不经用呢,正常使用下,还在使用其他模型的同时,几乎可以一天用掉一个 Codex 的重置周额度,2天左右也能够把他送的那个 62500 积分大概 2500 刀的 Credits 用完,甚至有一种边用你看到额度逐步下降的那种倒计时的感觉,很 Pro 不 Max。 真希望 OpenAI 加油💪 做到和 opus 5.5 一样的速度快、效果好、额度足的这不可能三角,也更希望国产模型加油,好久没有看到发布亮眼模型了。
16
Maybe it's just pea sized brain of mine, but I thought we all study "solved" problems since we are born? Plus nothing stops mathematicians from reinventing the wheel, just like js devs do! Sure it changes the landscape, and it already did for other professions.
BREAKING: Terence Tao and 24 other Fields Medalists just signed a letter telling AI companies they're destroying mathematics >headlines: AI solves famous math problems >25 greatest mathematicians respond today >"we are witnessing a general threat to intellectual work" >ai labs are treating math problems like a benchmark you can brute force >“solutions are announced in a rush, leaving no time for a proper writeup and citing relevant previous work of others" >they're talking about openai >"this raises severe attribution and plagiarism questions" >they're definitely talking about openai >openai offered to put math professor name on navier-stokes proof >condition: abandon his coauthor bc he works at anthropic >"the most precious resources of our profession are students and ideas" >for the labs the most precious resource is GPUs A Severe Misalignment of AI in Mathematics
1
1
33