kofdai

8 ms·

1.
Pure Code: What remains with developers after AI writes the code?(pure-code-essay.vercel.app)
4 points by kofdai 15d ago|0 comments
2.
Show HN: C. Build with AI. Keep the Experience(cleanroom-qvq1xbb8o-ag3497120s-projects.vercel.app)
1 points by kofdai 18d ago|0 comments
3.
by kofdai 1mo ago
Hello, hn I failed many times when trying to write in my diary every day. So, I decided to tie the timing of my entries not to my own will, but to physical events that are bound to happen.
4.
Show HN: Hitomaki – A diary that advances when you finish a toilet roll(ag3497120.github.io)
2 points by kofdai 1mo ago|1 comments
5.
by kofdai 1mo ago
What configuration did you use? I haven't been able to test it in a sufficiently long-term session yet, so I'll report back once I'm done.
6.
by kofdai 1mo ago
Hi HN, this works on a very simple mechanism. It’s an ultra-lightweight shared memory system that relies on modern AI’s ability to search through its memory, so it doesn’t even use similarity searches. The commands are simple, too. I built
7.
Show HN: The lightweight Claude and LMStudio, and memory across sessions.(call-me-vera.vercel.app)
3 points by kofdai 1mo ago|3 comments
8.
by kofdai 3mo ago
I refreshed the page, but it doesn't seem to have updated. I refreshed it again. The homepage used to show unrelated iOS games—the design is poor, but it should now display content related to this thread. Sorry for the inconvenience.
9.
Language vs. Vectors(verantyx.ai)
2 points by kofdai 3mo ago|2 comments
10.
Show HN:Swarm intelligence without degradation using two Qwen models
1 points by kofdai 3mo ago|0 comments
11.
by kofdai 5mo ago
Thank you for your feedback. I have reconsidered the structure and revised it as follows. Please take a look. ## The Heart of Gatekeeper: "6-Axis Three-Dimensional Cross Structure (JCross 6-Axis IR)" Gatekeeper is the "absolu
12.
by kofdai 5mo ago
Thanks for the heads-up! Since I'm new here, I didn't realize the formatting rules and ended up missing the 2-hour edit window. I've replied to my original comment with the properly formatted version. I appreciate the help! I
13.
Verantyx – A native IDE that obfuscates code before sending it to Cloud LLMs(github.com)
4 points by kofdai 5mo ago|5 comments
14.
by kofdai 5mo ago
I’m a student developer building Verantyx, a native macOS IDE solving a core dilemma in AI coding: We need the reasoning power of frontier models like Claude 4.7 Opus, but cannot legally or ethically leak proprietary source code, hardcoded
15.
by kofdai 7mo ago
Title: [Show HN] Verantyx Update: 22.7% on ARC-AGI-2 using Human-Logic + OpenClaw Loop Body: Following up on my previous post (where I was at 18.1%), I’ve just reached 22.7% (227/1000) on the ARC-AGI-2 public evaluation set. I want to
16.
by kofdai 7mo ago
Title: Clarification on my development workflow (re: jaen) You’re right to be skeptical of the speed, and I realize I was incomplete in describing my process. I should have been more transparent: I am using Claude Code as a "pair-progr
17.
by kofdai 7mo ago
You're right—I should have engaged with your actual point more carefully. Let me address it directly. You said the SotA is models generating code at test-time, and you're correct. Systems like o3 synthesize Python programs per-tas
18.
by kofdai 7mo ago
I understand the skepticism—the line between "AI-generated" and "AI-assisted" has become incredibly blurry. Let me clarify the architectural distinction. 1. The Inference Engine is 100% Deterministic: The "solver&qu
19.
I beat Grok 4 on ARC-AGI-2 using a CPU-only symbolic engine (18.1% score)(github.com)
11 points by kofdai 7mo ago|6 comments
20.
by kofdai 9mo ago
VERANTYX is an experiment in making AI reasoning structural rather than purely statistical. The system: • runs locally • uses small, explicit rule sets • falls back to LLMs only when structure is missing • marks all mined output as prov
21.
by kofdai 9mo ago
Hello HN, I've spent the last year building AXIS, a system that fundamentally reframes how we use language models in reasoning tasks. The core insight: LLMs are unreliable as reasoners , but powerful as knowledge extractors . So wh