Hacker News: codybontecou

New comment by codybontecou in "Show HN: Claude Code vs. Codex Global Usage Leaderboard"

codybontecou — Fri, 15 May 2026 20:12:31 +0000

Pay to publish your team's ai usage*

New comment by codybontecou in "Elevated error rates on Opus 4.7"

codybontecou — Fri, 15 May 2026 01:47:54 +0000

I tried to like it, but it eventually got stuck in a near-infinite loop trying to debug an extra curly bracket in an iOS app.

That and the lack of image-read support surprised me. I'm a big fan of feeding screenshots into my llm and that killed it for me.

New comment by codybontecou in "MIT: 20% drop in incoming graduate students"

codybontecou — Thu, 14 May 2026 18:51:40 +0000

Oh this is neat. Thank you.

New comment by codybontecou in "A message from President Kornbluth about funding and the talent pipeline"

codybontecou — Thu, 14 May 2026 16:52:57 +0000

What are some of the Chinese colleges worth paying attention to? Is most of the AI research coming from these colleges or are they primarily from private labs?

New comment by codybontecou in "EU to crack down on TikTok, Instagram's 'addictive design' targeting kids"

codybontecou — Tue, 12 May 2026 15:28:07 +0000

Which entrepreneurial sites are you browsing? I feel like I'm at a similar stage in my day-to-day.

New comment by codybontecou in "If AI writes your code, why use Python?"

codybontecou — Tue, 12 May 2026 14:09:03 +0000

You're devolving into the realm of "What if we tell the agent to just get it right?"

Relying on the prompt to ensure the code it writes is correct is where things fail. Types, tests, linting, etc. are deterministic tools the agents tend to respect.

Show HN: Iso.me – Open-source, on-device iOS location tracker

codybontecou — Thu, 30 Apr 2026 14:29:00 +0000

Article URL: https://github.com/CodyBontecou/isome

Comments URL: https://news.ycombinator.com/item?id=47963067

Points: 2

# Comments: 0

New comment by codybontecou in "The beginning of scarcity in AI"

codybontecou — Thu, 16 Apr 2026 22:29:01 +0000

pi vs. claude code vs. codex These are all agent harnesses which run a model (in pi's case, any model) with a system prompt and their own default set of tools.

New comment by codybontecou in "Google Gemma 4 Runs Natively on iPhone with Full Offline AI Inference"

codybontecou — Wed, 15 Apr 2026 15:43:09 +0000

It's a custom agent harness with on-device models and the ability to swap between models.

Basically, a "toy" app to showcase where we are with coding agents on-device.

New comment by codybontecou in "Google Gemma 4 Runs Natively on iPhone with Full Offline AI Inference"

codybontecou — Wed, 15 Apr 2026 15:41:52 +0000

They've been slowly cutting them off of updates and/or taking them off the app store entirely.

See Anywhere and Replit. Anywhere was the #1 or #2 app and was taken off the app store entirely before being put on and then taken off again.

Last I checked, Replit hasn't received an update on the iOS app store in over two months due to reviews denying them.

New comment by codybontecou in "Google Gemma 4 Runs Natively on iPhone with Full Offline AI Inference"

codybontecou — Wed, 15 Apr 2026 10:40:45 +0000

Unfortunately Apple appears to be blocking the use of these llms within apps on their app store. I've been trying to ship an app that contains local llms and have hit a brick wall with issue 2.5.2

New comment by codybontecou in "Claude.ai down"

codybontecou — Mon, 13 Apr 2026 15:51:37 +0000

I think it's model-specific. Opus 4.6 was 500ing but Sonnet 4.6 (non-1m context limit) is working.

New comment by codybontecou in "ChatGPT Pro now starts at $100/month"

codybontecou — Thu, 09 Apr 2026 21:45:29 +0000

That's an odd choice. I wonder why.

New comment by codybontecou in "The cult of vibe coding is dogfooding run amok"

codybontecou — Mon, 06 Apr 2026 19:17:28 +0000

Even if AI progress plateaus, I'm confident we would build tooling and patterns around the current models that would surpass hand crafted equivalents.

New comment by codybontecou in "Tell HN: Anthropic no longer allowing Claude Code subscriptions to use OpenClaw"

codybontecou — Sat, 04 Apr 2026 00:54:56 +0000

Are you using the Chinese models through their individual services or via an intermediary layer?

New comment by codybontecou in "Show HN: Sheet Ninja – Google Sheets as a CRUD Back End for Vibe Coders"

codybontecou — Sun, 29 Mar 2026 12:16:25 +0000

Most are lucky to get a few sign ups.

New comment by codybontecou in "Schedule tasks on the web"

codybontecou — Fri, 27 Mar 2026 22:07:27 +0000

Don't bet against the models and their providers becoming stagnant. Build with the idea that they will continue to improve.

New comment by codybontecou in "Thoughts on slowing the fuck down"

codybontecou — Thu, 26 Mar 2026 00:35:01 +0000

How did you find patterns between these sentences?

New comment by codybontecou in "So where are all the AI apps?"

codybontecou — Tue, 24 Mar 2026 14:44:48 +0000

Stuck behind Apple's app review process.

New comment by codybontecou in "Show HN: Crust – A CLI framework for TypeScript and Bun"

codybontecou — Tue, 17 Mar 2026 21:09:32 +0000

That’s… actually a great definition. I’m going to try to retain that.