Hacker News: boredumb

New comment by boredumb in "Redesigning source control at Anthropic"

boredumb — Tue, 13 Oct 2026 20:10:13 +0000

maybe next month the agents will learn the 'commit locally' and 'git squash' skills

New comment by boredumb in "On caring for user data: NeoVim caused Vim undo files to be deleted"

boredumb — Tue, 29 Sep 2026 17:07:39 +0000

You can pay for google docs+ via workspaces and they have an explicit customer data retention guarantee in their license.

New comment by boredumb in "On caring for user data: NeoVim caused Vim undo files to be deleted"

boredumb — Sun, 27 Sep 2026 16:01:34 +0000

You buy software that includes a commercial license that guarantees data retention

New comment by boredumb in "Jev in 25 Lines of Python"

boredumb — Wed, 23 Sep 2026 13:15:27 +0000

> LLMs are surprisingly biased towards picking "A", especially if they're otherwise not sure.

Not nearly as sophisticated as myself who would mutter "When in doubt - Charlie out" before marking C.

New comment by boredumb in "MCP was always a bad idea?"

boredumb — Mon, 21 Sep 2026 15:55:12 +0000

Just have a route that describes your API endpoints in a machine readable format, eventually they'll come to some consensus and in the mean time this works well enough to allow LLMs to plan.

New comment by boredumb in "Pirate Face Rescues LLM Models from Deletion"

boredumb — Sun, 20 Sep 2026 17:07:51 +0000

ages ago I tried using IPFS to more or less accomplish this, I imagined it to act more like a weights/training data network fs that everyone would be able to participate in.

New comment by boredumb in "Astra for Law"

boredumb — Fri, 18 Sep 2026 10:45:17 +0000

What happens if the gun is faulty and explodes killing the guy next to me and the family wants compensation?

New comment by boredumb in "Astra for Law"

boredumb — Fri, 18 Sep 2026 08:52:45 +0000

In the US at least if a lawyer screws up a contract beyond what a competent lawyer would and causes financial harm they can be sued.

New comment by boredumb in "Astra for Law"

boredumb — Thu, 17 Sep 2026 20:39:03 +0000

memes and snark aside, can you use this to create legitimate terms and contracts for my products and if I do who is getting sued when it is wrong?

New comment by boredumb in "Show HN: Craigslist for agent skills, curated by a human"

boredumb — Thu, 17 Sep 2026 20:04:08 +0000

AIngies list

New comment by boredumb in "Ask HN: How to start a sales and marketing engine?"

boredumb — Thu, 17 Sep 2026 15:13:35 +0000

well, that's about it for the thread I imagine :D

thanks

New comment by boredumb in "AWS says it can't restore some data from mideast facilities struck by Iran"

boredumb — Thu, 17 Sep 2026 15:10:58 +0000

funny enough, there have been some huge problems involving congressman with both bathrooms and children in regards to moral diligence.

Ask HN: How to start a sales and marketing engine?

boredumb — Thu, 17 Sep 2026 14:22:55 +0000

Outside of personal outreach and marketing, what are the best avenues for jumpstarting a bootstrapped product into the market without an existing sales and marketing engine?

Selling/being acquired by a larger firm with the resources, trying to secure grants, looking for a VC/partner to invest, or some other secret sauce?

Comments URL: https://news.ycombinator.com/item?id=49741227

Points: 1

# Comments: 2

When AI Safety become market capture

boredumb — Sat, 12 Sep 2026 20:30:05 +0000

The danger is not that AI makes yesterday’s work easier or kills humanity. The danger is that we use it only to produce more of yesterday’s work. more ads, more low qual content, more bureaucratic language, more basura.

Meanwhile, some of the loudest voices selling “AI safety” deserve antitrust scrutiny. When incumbent firms warn that the technology is too dangerous for ordinary competition but conveniently safe inside their own products, partnerships, and forthcoming IPOs...fear begins to look less like public service and more like market capture. Safety rules should if anything try to address measurable harm and not this cycles IPO no matter how many people cry in front of the media.

I know astra and claude make banging out the boiler plate nice enough to mallet into production but they aren't building trusted mechanisms and the meme they are about to transcend humanity based on cloning hugging face repositories and come kill me with my home security or destroy humanity because fiverr.com is being gutted right before a historic financial event should be scrutinized more than "i like x model more because i did a prompt that worked" it should be treated like financial fraud, these people are telling normal people they are going to die if they aren't funded and supported and allowed to gate keep 3rd parties from participating, how is this happening in the USA? I don't want to hear trump I want to hear how this is remotely legal for anyone to threaten the widespread death and destruction of humanity via surrogate vested participants if the alternative is not trusting them with their livelihoods and physical security?

Beyond offensive this, in my opinion, is illegal activity that should be absolutely destroyed and not by pushing dog shit legislation that trips up normal people building AI on any frontiers but these people from lording over non-technical people with literal death threats like they are working on the manhattan project when they are working on an interactive search engine.

Comments URL: https://news.ycombinator.com/item?id=49676869

Points: 1

# Comments: 1

New comment by boredumb in "Ask HN: Who is using MCP in production?"

boredumb — Fri, 04 Sep 2026 12:15:21 +0000

We offer MCP and then consume it with our in-app assistant to go from a non technical prompt to a series of what is essentially API calls they can automate for themselves for repetitive tasks or things that require a few screens to accomplish can be done from the assistant widget itself, etc.

New comment by boredumb in "Was modern art a CIA psy-op? (2020)"

boredumb — Tue, 25 Aug 2026 12:33:58 +0000

Let's hear what non-default country invented pop-art in the 50s.

New comment by boredumb in "I were 17, I'd learn how to build LLMs from scratch"

boredumb — Mon, 24 Aug 2026 10:45:07 +0000

I agree with mostly all of this, but personally I wrote a toy LLM almost 5 years ago and while it never saw much use outside of boring my wife with a shitty command line demo with glee it did help me understand how they worked and how to apply them, played a lot with JAX and pytorch, ended up building a ghetto version of MCP and an LLM-Pool to proxy requests to my baby local models and so I didn't struggle to see the evolution of openrouter and MCP agentic workflows. The same way i'm really glad when I was younger I built a bad webserver by myself, a really painful SQLx type database, etc etc etc - none of these things led me to developing for Nginx or Oracle nor will knowing JAX get me a job at an AI research lab, but I do have a lot of depth in understanding how the technology works so that the flavors on top of them are easy to digest and make more use of immediately, and I think the same can be said for engineers coming into the field - if it's a spooky LLM box you aren't going to be squeezing the same amount of juice as the guy that knows how they work inside and out so having at least the understanding of a _babys first LLM_ is going to get you miles ahead of people who don't.

For anyone who wants to dork around there is https://github.com/rasbt/LLMs-from-scratch which is something amazing that I think anyone who wants to engineer things around LLMs should at least blast through and read.

New comment by boredumb in "Anthropic appears to be A/B testing reduced effort levels in Claude Code"

boredumb — Sat, 22 Aug 2026 19:08:43 +0000

I'm not worried about the volatility in the definition, i'm worried that I give it 1 token today and receive 2 token output, tomorrow I receive 40. If i'm doing this a hundred thousand times a day it is difficult to price this in for users downstream or in the extreme cases be able to absorb that at all short of going into a failmode with degraded access until someone goes and buys more tokens or gets the bill. The alternative is just pass the buck and bill your non-technical customers with a "tokens" line iteim every month.

New comment by boredumb in "New MCP Roadmap"

boredumb — Sat, 22 Aug 2026 18:28:32 +0000

I see. In my head it would be something like the agents harness having a list of services it interacts with, reaches out to service.com/agents.md for a fresh copy every so often and uses that to resolve the relevant tool calls.

New comment by boredumb in "Anthropic appears to be A/B testing reduced effort levels in Claude Code"

boredumb — Sat, 22 Aug 2026 18:16:00 +0000

there are roughly three of them and they all use the same pricing model. I am also not in the position to build a frontier model company these days.