Hacker News: messh

New comment by messh in "Ideas on modernizing the open-source desktop"

messh — Thu, 24 Sep 2026 14:58:42 +0000

I use a Mac at work and a pc w/vanilla KDE at home. And... I don't feel anything is less good with KDE? If anything I prefer that over modern Mac

New comment by messh in "Show HN: Drop – A rootless Linux sandbox with gVisor support"

messh — Tue, 22 Sep 2026 15:19:17 +0000

How is this different than bwrap or srt and others? Im using bwrap to achieve read only everywhere and and write on pwd. Also pi and other coding agents all have sandboxing that work in similar way

New comment by messh in "Show HN: Lossless-memory – a personal AI memory that never summarizes"

messh — Mon, 21 Sep 2026 21:51:45 +0000

If there are no summaries then when context is full messages need to get evicted. If doing so one by one then it would indeed destroy the cache. Of course... maybe the implementation evicted 50% of messages at once, I didnt verify in code

New comment by messh in "OpenJev"

messh — Fri, 18 Sep 2026 15:28:17 +0000

In Jev you pass options in the input and its output just gives some probability for each. Oai structured output just follows a schema. The exact output is still generated and there is no probability

New comment by messh in "Show HN: Capsule – Single-file web apps that save their data into SQLite"

messh — Tue, 15 Sep 2026 17:45:11 +0000

this is for local apps

New comment by messh in "Pion, an agent designed to run any company autonomously"

messh — Mon, 14 Sep 2026 22:43:35 +0000

how do you decide to start a new agent, and when to kill? also... do you use some works-tealing style task board, or otherwise how would the agents get new tasks.

New comment by messh in "Txt: A fast, keyboard-driven terminal text editor for engineers"

messh — Fri, 11 Sep 2026 22:30:47 +0000

This is gonna be a hard name to google :)

Amazon Quick is now generally available on desktop

messh — Thu, 10 Sep 2026 18:56:26 +0000

Article URL: https://aws.amazon.com/blogs/machine-learning/amazon-quick-is-now-generally-available-on-desktop/

Comments URL: https://news.ycombinator.com/item?id=49648650

Points: 2

# Comments: 0

New comment by messh in "Muse – Meta’s personal AI agent"

messh — Tue, 08 Sep 2026 20:27:25 +0000

"A n̈ew kind of AI"... yeah... join the pile :)

New comment by messh in "I-have-ADHD: A skill to stop coding agents from burying the answer"

messh — Tue, 08 Sep 2026 15:47:31 +0000

good skill. but why should all this repo e needed for a small MD file??

New comment by messh in "Coop – Isolated VM Environments for Running Claude Code and Codex"

messh — Mon, 07 Sep 2026 20:46:23 +0000

How is this better than say using bubblewrap sandbox solutions. Is it safer, is that the advantage?

New comment by messh in "Elevated Errors for Multiple Models"

messh — Thu, 03 Sep 2026 15:02:19 +0000

What about AWS Bedrock??

New comment by messh in "Stop Making TUIs"

messh — Sat, 22 Aug 2026 15:12:10 +0000

TUIs also live where the rest of the interesting stuff happens, in the terninal. They don't force you to move to a different env

New comment by messh in "Stop Making TUIs"

messh — Sat, 22 Aug 2026 15:09:08 +0000

If you use the terminal a lot then TUIs keep you there wo the need to manage yet a other window in the OS. For e.g. you can run nvim in another pane in tmux.

New comment by messh in "Stop Making TUIs"

messh — Sat, 22 Aug 2026 14:55:21 +0000

Html has also some of these properties, targeting the browser. Don't get me wrong. I do like TUIs, they are light weight and compose very nicely with the rest of the terminal (eg tmux). Also, i generally prefer just cli commands over tuis when possible.

New comment by messh in "Show HN: Huzzah – a novel approach to coding with AI"

messh — Fri, 21 Aug 2026 00:22:28 +0000

The downside is that... you don't get to brainstorm with the agent about ways to implement stuff. Architecture, design etc. It would feel like.. you're on your own. Of course you could just a regular chat interface alongside but theb you deal with two interfaces. Maybe this could be a target format for another chat agent that you talk to.

New comment by messh in "fx :Tiny, open, native coding agent."

messh — Thu, 20 Aug 2026 15:01:54 +0000

For eg. on a large codebase I do hierarchical review where higher level agents focus on modules and leafs on files. Each reports upwards a summary of its findings incl children summaries.

New comment by messh in "fx :Tiny, open, native coding agent."

messh — Thu, 20 Aug 2026 03:15:21 +0000

that's an amazing project! people always say why 6mb vs CC's 250mb even matter when you are calling out to LLMs hosted in the cloud. But... I regularly run hierarchies of agents with say 50-100 on a regular basis. So 650 vs 25050 ... is "can do" vs "cannot"

New comment by messh in "My Agent Setup"

messh — Thu, 13 Aug 2026 15:06:31 +0000

The x2 instance has 8gb memory. They go up to x8. And if you don't use it say during a weekend, then you stop it and resume on Monday, not paying for that period

New comment by messh in "Show HN: Needle2: 14MB agentic LLM for phones, wearables, smart home and robots"

messh — Tue, 11 Aug 2026 20:35:49 +0000

For the microcm, take a look at https://shellbox.dev, se3ms like a good fit. Real linux, and very affordable