Hacker News: SparkyMcUnicorn

New comment by SparkyMcUnicorn in "System Card: Claude Mythos Preview [pdf]"

SparkyMcUnicorn — Tue, 07 Apr 2026 23:34:17 +0000

https://marginlab.ai/trackers/claude-code-historical-perform...

New comment by SparkyMcUnicorn in "April 2026 TLDR Setup for Ollama and Gemma 4 26B on a Mac mini"

SparkyMcUnicorn — Fri, 03 Apr 2026 15:08:11 +0000

I've found my research on certain topics like this becoming less reliable these days, compared to just trying it out to form an opinion.

New comment by SparkyMcUnicorn in "Show HN: ctx – an Agentic Development Environment (ADE)"

SparkyMcUnicorn — Fri, 03 Apr 2026 14:54:32 +0000

Your repo says it's open source, but it's missing the source.

https://github.com/ctxrs/ctx

New comment by SparkyMcUnicorn in "I built an AI receptionist for a mechanic shop"

SparkyMcUnicorn — Mon, 23 Mar 2026 21:09:46 +0000

I don't think it's a bug, it's an extra hoop to jump through.

New comment by SparkyMcUnicorn in "Show HN: Tmux-IDE, OSS agent-first terminal IDE"

SparkyMcUnicorn — Wed, 18 Mar 2026 19:19:23 +0000

It sounds like maybe you haven't seen agent teams mode, which this project is using.

https://code.claude.com/docs/en/agent-teams

New comment by SparkyMcUnicorn in "Mouser: An open source alternative to Logi-Plus mouse software"

SparkyMcUnicorn — Fri, 13 Mar 2026 20:59:56 +0000

I use the g305, and have to replace it roughly once a year because the scroll button eventually stops working. I've been through 5-6 of them. Regularly on sale for around $25.

Wish I could find a better mouse I like as much as this one that I didn't need to pay a yearly subscription for. It's just the right size, lightweight, wireless, and being able to store the customizations to on board memory is nice.

New comment by SparkyMcUnicorn in "Mullvad VPN: Banned TV Ad in the Streets of London [video]"

SparkyMcUnicorn — Thu, 05 Mar 2026 00:07:34 +0000

> By providing an example of similar articles about other locations I was showing that this alone is not particularly strong evidence.

How does linking similar articles help demonstrate that? That doesn't seem like very strong (counter?)evidence.

New comment by SparkyMcUnicorn in "Mullvad VPN: Banned TV Ad in the Streets of London [video]"

SparkyMcUnicorn — Tue, 03 Mar 2026 17:12:01 +0000

This is whataboutism. It doesn't address any of the points made above.

New comment by SparkyMcUnicorn in "Halt and Catch Fire: TV’s best drama you’ve probably never heard of (2021)"

SparkyMcUnicorn — Wed, 18 Feb 2026 04:39:40 +0000

Somewhat mysteriously, the Linux ISOs for this show have seen a sudden spike in activity.

New comment by SparkyMcUnicorn in "An Update on Heroku"

SparkyMcUnicorn — Fri, 06 Feb 2026 16:08:03 +0000

> lead to customers moving elsewhere

Since they're no longer accepting new enterprise clients, maybe this is intentional.

New comment by SparkyMcUnicorn in "Netbird – Open Source Zero Trust Networking"

SparkyMcUnicorn — Sun, 01 Feb 2026 18:14:55 +0000

I've done this a few different ways.

I started with a docker container that connected to both the VPN provider and tailscale. Now OPNSense is handing a few connections to the VPN provider at a couple locations around the world, and enforcing external traffic to be routed to the VPN connections via VLAN tags (untagged has direct internet access).

Using the VPN provider can either be adding a VLAN tag to a machine/container or connecting to a "vpn-{location}" tailscale exit node.

New comment by SparkyMcUnicorn in "Show HN: macOS menu bar app to track Claude usage in real time"

SparkyMcUnicorn — Fri, 09 Jan 2026 00:58:23 +0000

Have you seen aichat[0]?

Doesn't support the "from anywhere" part, but the resume strategies are pretty cool.

[0] https://github.com/pchalasani/claude-code-tools#aichat-sessi...

New comment by SparkyMcUnicorn in "Show HN: Local Privacy Firewall-blocks PII and secrets before ChatGPT sees them"

SparkyMcUnicorn — Thu, 11 Dec 2025 23:14:51 +0000

I configure permission settings within projects.

https://code.claude.com/docs/en/settings#permission-settings

New comment by SparkyMcUnicorn in "GPT-5.2"

SparkyMcUnicorn — Thu, 11 Dec 2025 18:51:52 +0000

If the pretraining rumors are true, they're probably using continued pretraining on the older weights. Right?

New comment by SparkyMcUnicorn in "Cloudflare Sandbox SDK"

SparkyMcUnicorn — Thu, 16 Oct 2025 22:50:58 +0000

To me, the docs answer it pretty clearly. The defined directories persist until you destroy().

The part that's unclear to me is how billing works for a sandbox's disk that's asleep, because container disks are ephemeral and don't survive sleep[2] but the sandbox pricing points you to containers which says "Charges stop after the container instance goes to sleep".

https://developers.cloudflare.com/sandbox/concepts/sandboxes...

[2] https://developers.cloudflare.com/containers/faq/#is-disk-pe...

New comment by SparkyMcUnicorn in "Claude Code 2.0"

SparkyMcUnicorn — Tue, 30 Sep 2025 05:30:04 +0000

If you turn off "Help improve Claude" you will never get this prompt (I never do).

https://claude.ai/settings/data-privacy-controls

New comment by SparkyMcUnicorn in "Context is the bottleneck for coding agents now"

SparkyMcUnicorn — Fri, 26 Sep 2025 17:08:38 +0000

> For finicky issues like that I often find that, in the time it takes to create a prompt with the necessary context, I was able to just make the one line tweak myself.

I don't run into this problem. Maybe the type of code we're working on is just very different. In my experience, if a one-line tweak is the answer and I'm spending a lot of time tweaking a prompt, then I might be holding the tool wrong.

Agree on those terms being relative. Maybe a better way of putting it is that I'm very comfortable putting my name on it, deploying to production, and taking responsibility for any bugs.

New comment by SparkyMcUnicorn in "Context is the bottleneck for coding agents now"

SparkyMcUnicorn — Fri, 26 Sep 2025 15:36:13 +0000

I'll stop you right there. I've been using Claude Code for almost a year on production software with pretty large codebases. Both multi-repo and monorepo.

Claude is able to create entire PRs for me that are clean, well written, and maintainable.

Can it fail spectacularly? Yes, and it does sometimes. Can it be given good instructions and produce results that feel like magic? Also yes.

New comment by SparkyMcUnicorn in "Show HN: The text disappears when you screenshot it"

SparkyMcUnicorn — Thu, 18 Sep 2025 17:28:22 +0000

Yes: https://jsfiddle.net/kx6stbcL/

New comment by SparkyMcUnicorn in "Microsoft Favors Anthropic over OpenAI for Visual Studio Code"

SparkyMcUnicorn — Tue, 16 Sep 2025 21:21:45 +0000

I decided to give perplexity another try a few days ago, and it still seems to hallucinate things. Given the same exact tasks/prompts both Claude and Chatgpt got the facts correct.