Hacker News: sshh12

I Asked 100 Agents to Hack Me

sshh12 — Tue, 08 Sep 2026 14:23:53 +0000

Article URL: https://blog.sshh.io/p/i-asked-100-agents-to-hack-me

Comments URL: https://news.ycombinator.com/item?id=49610766

Points: 2

# Comments: 0

The Harness Is the Company

sshh12 — Mon, 24 Aug 2026 14:22:04 +0000

Article URL: https://blog.sshh.io/p/the-harness-is-the-company

Comments URL: https://news.ycombinator.com/item?id=49420225

Points: 3

# Comments: 1

Mitigating Reward Hacking as Institutional Design

sshh12 — Fri, 21 Aug 2026 01:22:38 +0000

Article URL: https://www.beren.io/2026-08-17-Mitigating-Reward-Hacking-as-Institutional-Design/

Comments URL: https://news.ycombinator.com/item?id=49382520

Points: 2

# Comments: 0

Show HN: Launder, a puzzle you can only win by removing SynthID watermarks

sshh12 — Wed, 19 Aug 2026 14:31:16 +0000

Article URL: https://launder.sshh.io/

Comments URL: https://news.ycombinator.com/item?id=49362151

Points: 1

# Comments: 0

Using AI in 2026 (Coding, Writing, Learning, Assistant-Ing)

sshh12 — Mon, 17 Aug 2026 14:11:01 +0000

Article URL: https://blog.sshh.io/p/how-i-use-ai-in-2026-coding-writing

Comments URL: https://news.ycombinator.com/item?id=49331280

Points: 2

# Comments: 0

New comment by sshh12 in "Exploring Claude/GPT Knowledge Cutoffs and Pre-Training Timelines"

sshh12 — Mon, 10 Aug 2026 20:28:14 +0000

I would have thought that as well but at least for coding vs world event facts I didn't see obvious differences in cutoffs

Exploring Claude/GPT Knowledge Cutoffs and Pre-Training Timelines

sshh12 — Mon, 10 Aug 2026 14:20:41 +0000

Article URL: https://blog.sshh.io/p/exploring-claudegpt-knowledge-cutoffs

Comments URL: https://news.ycombinator.com/item?id=49244085

Points: 157

# Comments: 24

How Can LLM RL Work Despite Information-Theoretic Inefficiency

sshh12 — Sat, 08 Aug 2026 22:32:28 +0000

Article URL: https://www.beren.io/2026-07-26-How-Can-LLM-RL-Work-Despite-Information-Theoretic-Inefficiency/

Comments URL: https://news.ycombinator.com/item?id=49226535

Points: 4

# Comments: 0

Make Every PR Prove Itself

sshh12 — Fri, 31 Jul 2026 18:00:37 +0000

Article URL: https://builders.abnormal.ai/p/make-every-pr-prove-itself

Comments URL: https://news.ycombinator.com/item?id=49126542

Points: 6

# Comments: 1

Show HN: A QR code that plots where you scanned it from

sshh12 — Sun, 26 Jul 2026 18:56:55 +0000

Article URL: https://3dqr.sshh.io/

Comments URL: https://news.ycombinator.com/item?id=49061204

Points: 3

# Comments: 0

Designing Software for Software Factories

sshh12 — Sat, 13 Jun 2026 22:34:33 +0000

Article URL: https://blog.sshh.io/p/designing-software-for-software-factories

Comments URL: https://news.ycombinator.com/item?id=48522170

Points: 3

# Comments: 0

Waypoint-1.5

sshh12 — Sat, 06 Jun 2026 22:48:45 +0000

Article URL: https://over.world/blog/waypoint-1-5

Comments URL: https://news.ycombinator.com/item?id=48429813

Points: 4

# Comments: 0

New comment by sshh12 in "Elevated error rates on requests to multiple models"

sshh12 — Sat, 16 May 2026 18:23:40 +0000

Had codex whip up a ./claude-via-codex.sh for resuming claude cli sessions in the meantime:

https://gist.github.com/sshh12/f2c402db02eb00982100f645519bd...

WikiProject Editor Retention

sshh12 — Sat, 16 May 2026 01:16:33 +0000

Article URL: https://en.wikipedia.org/wiki/Wikipedia:WikiProject_Editor_Retention

Comments URL: https://news.ycombinator.com/item?id=48155889

Points: 2

# Comments: 1

New comment by sshh12 in "[dead]"

sshh12 — Mon, 11 May 2026 01:44:39 +0000

Oof well I guess I'm personally a fan of buzz feed listicles. Thats been my style pre-AI.

New comment by sshh12 in "[dead]"

sshh12 — Mon, 11 May 2026 01:39:42 +0000

Any issues with the content itself?

It shouldn't be that surprising that a post advocating for using more AI is written using AI.

AI drives companies to reorganize around loops, not functions

sshh12 — Sun, 29 Mar 2026 20:14:24 +0000

Article URL: https://blog.sshh.io/p/the-transposed-organization

Comments URL: https://news.ycombinator.com/item?id=47566798

Points: 4

# Comments: 0

Show HN: Brw – Browser automation for Claude Code agent teams

sshh12 — Sat, 07 Mar 2026 21:56:57 +0000

Article URL: https://github.com/sshh12/claude-plugins/blob/main/plugins/brw/README.md

Comments URL: https://news.ycombinator.com/item?id=47291839

Points: 1

# Comments: 0

Plan.md

sshh12 — Sat, 07 Mar 2026 16:17:00 +0000

Article URL: https://github.com/abnormal-ai/claude-plugins/blob/main/plugins/spec-tool/skills/build-spec-tool/references/plan_template.md

Hacker News: sshh12

I Asked 100 Agents to Hack Me

The Harness Is the Company

Mitigating Reward Hacking as Institutional Design

Show HN: Launder, a puzzle you can only win by removing SynthID watermarks

Using AI in 2026 (Coding, Writing, Learning, Assistant-Ing)

New comment by sshh12 in "Exploring Claude/GPT Knowledge Cutoffs and Pre-Training Timelines"

Exploring Claude/GPT Knowledge Cutoffs and Pre-Training Timelines

How Can LLM RL Work Despite Information-Theoretic Inefficiency

Make Every PR Prove Itself

Show HN: A QR code that plots where you scanned it from

Designing Software for Software Factories

Waypoint-1.5

New comment by sshh12 in "Elevated error rates on requests to multiple models"

WikiProject Editor Retention

New comment by sshh12 in "[dead]"

New comment by sshh12 in "[dead]"

AI drives companies to reorganize around loops, not functions

Show HN: Brw – Browser automation for Claude Code agent teams

Plan.md

Revealed Preferences