Hacker News: hbarka

New comment by hbarka in "The new rules of context engineering for Claude 5 generation models"

hbarka — Sun, 26 Jul 2026 23:35:38 +0000

> The new rules of context engineering for Claude 5 generation models We removed over 80% of Claude Code's system prompt for more advanced models.

This is incredible. It’s like Claude learned one of the central ideas in the Tao Te Ching: you remove something every day. Wisdom and wholeness do not necessarily come from accumulating more. They can come from removing.

New comment by hbarka in "Google Discloses $94.1B in SpaceX Stock, Marking 6% Stake"

hbarka — Sun, 26 Jul 2026 20:46:57 +0000

Putting the principal idea of “publicly owned” in its head.

New comment by hbarka in "Jimothy the raccoon has a rare spinal condition. Here's what that means"

hbarka — Sun, 26 Jul 2026 18:51:53 +0000

I know that one. It’s the good ol’ tolerance formula:

Tolerance(A) = { 100% if A ≠ B; 0% if A → B }

Sprinkle on anything to suit your taste.

New comment by hbarka in "Jimothy the raccoon has a rare spinal condition. Here's what that means"

hbarka — Sun, 26 Jul 2026 18:35:30 +0000

God forbid we put out a pan of dry kibbles for a stray cat or a water pan outside for hiking dogs during hot days. That would be “enabling”.

New comment by hbarka in "How to Write English Prose"

hbarka — Sun, 26 Jul 2026 18:31:19 +0000

> A good part of our sleepes is peeced out with visions,

Because of current slang, I read this as “peaced out” then realized it’s “pieced out”. Peace out

New comment by hbarka in "NYC may require landlords and realtors to disclose the use of AI in listings"

hbarka — Sun, 19 Jul 2026 01:56:47 +0000

> and I cannot think of a logical reason to not provide that information.

The reason is simple. Omission is deception.

New comment by hbarka in "We're extending access to Fable 5 on all paid plans through July 12"

hbarka — Tue, 07 Jul 2026 18:36:22 +0000

Looks like we are at the end of the frontier-level models at subscription pricing. After this grace period, it will be double the cost at paid-per-token usage. I’m counting on the other models to compete at the subscription level and I need my harness to be agnostic. I need an AI harness that lets me switch LLM models dynamically depending on the task.

Ask HN: What is your AI harness that lets you switch LLM models easily?

hbarka — Tue, 07 Jul 2026 18:28:53 +0000

Looks like we are at the end of the Fable-level model at subscription pricing. It will be double the cost at paid-per-token usage. I’m counting on the other models to compete now at the subscription level and I need my harness to be agnostic. What do you use?

Comments URL: https://news.ycombinator.com/item?id=48821642

Points: 10

# Comments: 10

New comment by hbarka in "Atomic Force Microscope [video]"

hbarka — Sun, 05 Jul 2026 16:05:15 +0000

That video is a perfect description. Hard disk drives are incredible machines, going through many steps to produce the platter. The flying head is also fascinating. To this day I have full working hard drives over over twenty years old, while my SSDs have gone kaput.

The Bay Area used to be the center of these companies’ manufacturing and I was there when they literally unbolted the Varian sputter machines from my company to ship to Malaysia. The likes of Seagate, Western Digital, Conner, Quantum, Maxtor, Micropolis, Hitachi, Fujitsu, IBM, Read-Rite. Also Iomega in Utah. They were all here. They all left practically overnight. They were executive decisions. It was a shocking move and the whole history would make a good documentary or movie.

New comment by hbarka in "Atomic Force Microscope [video]"

hbarka — Sun, 05 Jul 2026 08:10:32 +0000

In my first internship, with a hard disk drive company, I learned how to use an Atomic Force Microscope to measure the roughness of the hard drive platter (the disk). The texture variation is in the order of angstroms or nanometers. It’s incredible how the AFM works like the needle of a record player, not via optics, and sensing at the atomic level.

A 'new way of thinking' boosted math proficiency in an East Palo Alto school

hbarka — Thu, 02 Jul 2026 07:17:38 +0000

Article URL: https://www.paloaltoonline.com/east-palo-alto/2026/07/01/a-new-way-of-thinking-boosted-math-proficiency-in-an-east-palo-alto-school/

Comments URL: https://news.ycombinator.com/item?id=48757684

Points: 10

# Comments: 1

New comment by hbarka in "Tokyo has only two barley tea makers, we visited one to see how mugicha is made"

hbarka — Tue, 30 Jun 2026 22:38:44 +0000

Also called boricha on Korean. Can find in local H Mart.

I have not tried the Japanese roasted barley variety but houjicha, roasted green tea, is excellent and can find in Japantown.

New comment by hbarka in "Show HN: Recall – fully-local project memory for Claude Code"

hbarka — Sun, 21 Jun 2026 22:10:12 +0000

Willing to try this but the author missed that Claude also has memory.md

New comment by hbarka in "Ask HN: What technique do you use to make Claude Code deterministic?"

hbarka — Sat, 20 Jun 2026 16:45:06 +0000

I agree with your example about a deterministic answer but what I’m looking for is deterministic process. I seek the LLM’s opinion, not a boolean answer. For example, having an agentic skill or hook to do a SWOT analysis may one day (out of 1000 consistent days) result in the agent just produce S-W-O and no T in the process because it was simply context muddled that day.

Ask HN: What technique do you use to make Claude Code deterministic?

hbarka — Sat, 20 Jun 2026 16:05:24 +0000

I’m curious to hear from the HN community how you tamed this probabilistic non-deterministic genius generate a deterministic and repeatable result 100% percent of the time.

Comments URL: https://news.ycombinator.com/item?id=48610280

Points: 8

# Comments: 11

Ask HN: What do you do to make LLMs determine

hbarka — Sat, 20 Jun 2026 15:59:03 +0000

Comments URL: https://news.ycombinator.com/item?id=48610221

Points: 2

# Comments: 0

New comment by hbarka in "Shepherd's Dog: A Game by the Most Dangerous AI Model"

hbarka — Sat, 13 Jun 2026 06:41:56 +0000

That’s one tired sheepdog.

New comment by hbarka in "Statement on US government directive to suspend access to Fable 5 and Mythos 5"

hbarka — Sat, 13 Jun 2026 06:33:12 +0000

> This has been going on for decades

With statements like that, if it’s been going on that long then it’s either our culture and normal way of life or you’re on some QAnon cuckoo rabbit hole.

New comment by hbarka in "Statement on US government directive to suspend access to Fable 5 and Mythos 5"

hbarka — Sat, 13 Jun 2026 06:02:05 +0000

> We saw some of this behavior in the last administration too.

Can we stop with this bothsides-ing. The level of co-opting by this administration is unprecedented. There’s the strong-arming to get Intel equity stake, Nvidia/AMD revenue share, U.S. Steel golden share, Lithium Americas equity stake, Big Law pro bono pledges, TikTok forcible acquisition, Paramount-CBS-Skydance favor, it’s just unbelievable the stark use of power.

New comment by hbarka in "Apache Burr: Build reliable AI agents and applications"

hbarka — Wed, 10 Jun 2026 17:05:21 +0000

Is this comparable to https://dspy.ai/ ?