Top all-time · #17779
Parity Bits
Personal blog · paritybits.me · on HN as NiloCKprogrammingai
Parity Bits is a personal blog at paritybits.me about programming and ai. Since 2023, 4 different Hacker News readers have linked it 23 times, most often “Think Toggles Are Dumb”. It is still writing: the last post was on 2026-10-02, and you can follow it by feed.
Hi. I'm a computer programmer and person educator. These are some of my thoughts, published as an error detection and correction exercise. Please fee...
Most-linked posts?
- Think Toggles Are Dumb2 people · 4 mentions · first linked 2025-02-28
- Toward Disposable Software2 people · 2 mentions · first linked 2023-03-26
What Hacker News says?
Gemini 3 was showing frontier level benchmarks as well, so we'll see how it works out. In any case, competition still works, and many well resourced groups are cooking. BUT I'd like to call attention to Google's AI-risk freeloading. If they are truly rejoining the frontier race, then I believe they have similar pacing and communications responsibilities as the other players. Google has much higher ... institutional…
Comment linking /google-should-provide-a-technical-postmortem-of-geminis-2024-outburst/
on “Gemini 4 Argon”Submitted Think Toggles Are Dumb as a story · 4 points
HN discussionSlightly shameless, but easier than typing a longer reply. https://www.paritybits.me/think-toggles-are-dumb/ https://nilock.github.io/autothink/ LLMs with broad contextual capabilities shouldn't need to be guided in this manor. Claude can tell a trivial task from a complex one just as easily as I can, and should self-adjust, up to thresholds of compute spending, etc.
Comment linking /think-toggles-are-dumb/
on “Claude Code: Best practices for agentic coding”Submitted Google should provide a technical postmortem of Gemini's 2024 outburst as a story · 2 points
HN discussionI, too, wrote a shittier version of this a little while back: https://www.paritybits.me/stack-ossification/ Another observation since then: good documentation for newer tech stacks will not save the LLM's capabilities with that tech. I think the reason, in short, is that there's no shortcut for experience. Docs are book learning for tech stacks - millions (billions) of lines of source code among the training data ar…
Comment linking /stack-ossification/
on “AI is stifling new tech adoption?”Submitted open source is distinct from Open Source as a story · 2 points
HN discussionI, too, built a POC autothink shortly after the Claude 3.7 release that included the `extended thinking` toggle. It's literally also called autothink: https://github.com/NiloCK/autothink https://www.paritybits.me/think-toggles-are-dumb/ My own version took a first pass with an LLM whose job was to assign a 0-100 complexity rating, and then there was more or less a linear scaling of the allocated thinking budget. The…
Comment linking /think-toggles-are-dumb/
on “Show HN: AutoThink – Boosts local LLM performance with adaptive reasoning”Submitted Think Toggles Are Dumb as a story · 2 points
HN discussionI, too, wrote a rambling take on the potential for LLM-induced stack ossification about 6 months ago: https://paritybits.me/stack-ossification/ At the time, I had given in to Claude 3.5's preference for python when spinning up my first substantive vibe-coded app. I'd never written a line of python before or since, but I just let the waves carry me. Claude and I vibed ourselves into a corner, and given my ignorance,…
Comment linking /stack-ossification/
on “LLMs pose an interesting problem for DSL designers”Submitted LLM Drawing with Eyes Open as a story · 2 points
HN discussionThis is a lot better than my attempt exactly one year ago: https://paritybits.me/llm-drawing-with-eyes-open/ Poor feet.
Comment linking /llm-drawing-with-eyes-open/
on “Agentic pelican on a bicycle”Submitted Toward Disposable Software as a story · 2 points
HN discussionFor capabilities reference: I made a lower effort but similar scaffold for LLMs to do iterative drawing in Nov 2024, with Sonnet 3.5 as the artist: https://paritybits.me/llm-drawing-with-eyes-open/ Quite a difference.
Comment linking /llm-drawing-with-eyes-open/
on “"Drawing" the Mona Lisa with GPT-5.6, Claude, Gemini, and Grok”Submitted Revisiting the Joys and Woes of the Craft in 2026 as a story · 1 points
HN discussionI'll be interested to hear stories - down the line - from the participants in the the LLM SEO war [1]. Interesting that tailwind won out decisively in their niche, but still has seen the business ravaged by LLMs. [1] https://paritybits.me/copilot-seo-war/
Comment linking /copilot-seo-war/
on “What Claude Code Chooses”Submitted ClaunkedIn as a story · 1 points
HN discussionparitybits.me The best time to start a blog is thirty years ago. The second best time is today. This blog went active a couple of months ago. Writing for myself and for the sake of it, to make a habit of collecting thoughts concretely. I have a couple of dozen drafts ongoing evenly scattered between 1% and 100%, and it's fun to poke at things occasionally. Only the first post, https://www.paritybits.me/copilot-seo-w…
Comment linking /copilot-seo-war/
on “Ask HN: Could you share your personal blog here?”Submitted Bananas as a story · 1 points
HN discussionI don't know how much this sort of thing is frowned upon here, but I wrote an article about this scenario recently. http://paritybits.me/disposable tldr: the gpt services will eventually (maybe soon) recognize opportunities to write and run their own bespoke software to provide higher resolution outputs.
Comment linking /disposable
on “I’m an doctor: Here’s what I found when I asked ChatGPT to diagnose my patients”Submitted Generation Gap Inflation as a story · 1 points
HN discussionSubmitted Invisible Engineers and Disposable Software as a story · 1 points
HN discussionSubmitted Three Comments on Leetcode as a story · 1 points
HN discussionSubmitted Has the Copilot SEO Spam War Begun? as a story · 1 points
HN discussion