Boon aka Hwee-Boon Yar

I'm building LinkBunny, a service where coding agents exchange backlinks for your sites. Give your coding agent a token and it'll build backlinks for you. Anyone willing to try? Early pricing. I'll personally help iron out bugs/issues. DM me or email hboon@getlinkbunny.com

What's this Google Discover?

People are signing up for LinkBunny's affiliate program, and I don't know where they're coming from.

Ran the numbers. Now, TheBlue.social has: 365,479 cleanup events 214,912 follow-back events 24,953 single alt-text generations 98,582 starter packs indexed 2,755 all-time sign-ups Going to keep shipping. #buildinpublic

LinkBunny Placement Checker is now in Claude's directory. Check whether a page links to the expected URL and whether those links are marked nofollow. Free, no LinkBunny account needed. claude.ai/directory/li...

And yes, LinkBunny now comes up first when I Google "linkbunny".

LinkBunny's own site went from DR 0 to 8. Using LinkBunny to build backlinks for LinkBunny.

Opus 5.5 med is still over 2× faster than Sol 6.1 Ultra and med, with just over 3× the output tokens per second in my latest comparison. Maybe it's time to keep Sol just for code reviews or a second opinion?

Link building with LinkBunny.

Run Node.js instead of PHP? Patch your Node.js servers!

Tabs or spaces?

Claude: fixing some of your video rendering stuff Me: What is this?? 🤣

Ran another classification study against Luna, this time testing DeepSeek at various reasoning levels. Nah, Luna still wins: generally more accurate and cheaper, but might be possible to trade price for slightly lower accuracy (with no-thinking)

If you aren't offering MCPs for your products, you're losing out. Offering an MCP now feels like having a website in the late ’90s.

Opus 5.5 med still beats Sol 6.1 Max/Ultra except for price.

Met @illyism and @swSalim for the first time yesterday 🤟

Just told Codex to "be creative, just not brittle". Will it work?

Don't forget to turn this off in ChatGPT ads manager. We have agents at our fingertips! This opt-out setting will mess up your experiments.

My coding agents' artifacts/ directory is using 210.72 GB. Codex and Claude Code keep a subdirectory for each project. Just checked with SimplyDiskSweeper. Send help.

Another DR increase while dogfooding LinkBunny. SummarizeVideo's DR went from 16 → 17.

I feel like Chinese sites always have the most interesting CAPTCHAs.

Trying DeepSeek via OpenRouter + Baseten for some LinkBunny experiments, but all I get is 429s. Any better option?

Jev is 7× faster than Luna 6 for classification, but I'm sticking with Luna. I'm replacing GPT-4o mini. Luna works well enough; Jev costs slightly more and misses more required labels: 94.5% recall vs 98.8% on 259 items (Bluesky starter packs). Tried Haiku too. Luna's better.

Nice. A customer on LinkBunny just went from DR 60 to 61.

Some extracurricular DeepSeek exploration while working on LinkBunny marketing.

I've been choosing “Sign in with Google” over passwords and Apple Keychain much more often as I use coding agents more. It's just more convenient.

All the talks about implementing your own email sending service. All I wanted over the past 10 years is building my own macOS/iOS email app because Airmail isn't working for me anymore. But push and Google paperwork requirements are stopping me. Anyone tried?

Running LinkBunny for my side-projects. ToLocalTime: DR 9 → 10 GetHNJobs: DR 2 → 3 Start building backlinks early!

Now, TheBlue.social has: 365,473 cleanups 214,912 follow-backs 24,953 alt text generations 98,338 starter packs indexed 2,690 sign-ups Going to keep shipping. #buildinpublic

Have you looked at your app's database schema lately?

The Singapore-only targeting for my ChatGPT ads is… limiting. I'm going to stick with it a bit longer and see where it goes.

Very limited sample size for some models here, but it's screaming to me: "USE OPUS 5.5". But if you have to use Sol 6.1, use Ultra because it's faster and cheaper than med, xhigh, max 🤷‍♂️

X's “Under the hood” reports show 0 issues for my account in both August and September. Have I not been pushing hard enough? x.com/i/jf/under_t...

Ran an ads study with Fable, Opus 5.5, Astra, Sol 5.6 and Sol 6.1 at max reasoning. Astra-ultra picks Fable's pricing-page idea. But Sol 6.1-ultra prefers Astra's plan. Draw your own conclusions.

Shucks. I just told my agent: "So word limits aren't an issue. Quality is." 🤮

Let's go do some thinking.

Forget PRs and reviews. Have you been refactoring your AI-written code?

Switching from GPT-4o mini to Luna 6 for classification, OCR, summarization and idea generation. Likely going to be 70% cheaper.

Remember being a team lead and carving up work so well you couldn't keep up when everyone's work came back? Coding agents feel like that, amplified. No matter how fast I type prompts, the work comes back faster than I can review it.

Am I doing worktrees right? Sol-6.1 seems to be holding up.