Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
rcarmo
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
4 ms
·
31.
▲
by
rcarmo
1mo ago
Started reading, saw “The timing is the story”, and my LLM-generated-spider-sense tingled. The piece is informative, but I now have no idea if the arguments are human or AI driven
32.
▲
Gpt 5.6 Sol might be just a tad too proactive
(taoofmac.com)
1 points
by
rcarmo
1mo ago
|
0 comments
33.
▲
by
rcarmo
1mo ago
I'm doing both "documents" and documents (JSON and gzip compressed blobs for emails, office docs, etc.), with the plaintext extracted and run through FTS5. I can't think of another database that would let me do this, plu
34.
▲
AI Image Generation on an RP2350 Microcontroller
(cpldcpu.github.io)
2 points
by
rcarmo
1mo ago
|
1 comments
35.
▲
by
rcarmo
1mo ago
Well, “low” is relative. Not everyone can spend a small car’s worth on a Studio…
36.
▲
by
rcarmo
1mo ago
The big difference you're not looking at is the memory bandwidth. The M5 Pro has nearly twice the one of the M6: 170GB/s memory bandwidth Configurable up to 32GB unified memory 307GB/s memory bandwidth Configurable up to 64GB
37.
▲
Tiny World Model
(paul.mou.dev)
3 points
by
rcarmo
1mo ago
|
0 comments
38.
▲
A 2026 Survey of Rust GUI Libraries
(blog.wybxc.cc)
2 points
by
rcarmo
2mo ago
|
0 comments
39.
▲
Bluetooth Shock Collar Keeps Wearer on Task
(hackaday.com)
4 points
by
rcarmo
2mo ago
|
0 comments
40.
▲
by
rcarmo
2mo ago
This has been a problem for me and a few folk I know ever since layoffs became routine. Every corporate reorg became a trigger for internal downscrolling as well
41.
▲
by
rcarmo
2mo ago
Well, we did it for politicians. Can't get any worse.
42.
▲
by
rcarmo
2mo ago
As I was reading this, I realized that the ESP32 HomeKit debugging session I had running was taking a while, and this happened: Me: what are you doing? Agent: I’m resolving Apple’s private video-resolution lookup table to determine whether
43.
▲
Apple Papercuts Summer Edition
(taoofmac.com)
1 points
by
rcarmo
2mo ago
|
0 comments
44.
▲
Building Piclaw on Top of an Opinionated Coding Agent
(taoofmac.com)
4 points
by
rcarmo
2mo ago
|
0 comments
45.
▲
by
rcarmo
2mo ago
This is lovely and a reminder of why I used pi as a basis for my own harness. I have a similar idea in LISP that isn’t as minimal, might go all in on eval…
46.
▲
Frasier Fantasy: The Director's Cut by Edward La Barbera
(edward-la-barbera.itch.io)
2 points
by
rcarmo
2mo ago
|
0 comments
47.
▲
by
rcarmo
2mo ago
Delightful, even if I don’t recall Solaris
48.
▲
Rcarmo/bun-packrat: Single-file web archive: SQLite-backed, self-contained HTML
(github.com)
1 points
by
rcarmo
2mo ago
|
0 comments
49.
▲
Late-night-regrets – piclaw addon for agent self-correction
(rcarmo.github.io)
1 points
by
rcarmo
2mo ago
|
0 comments
50.
▲
3D Printing State of the Union
(taoofmac.com)
1 points
by
rcarmo
2mo ago
|
0 comments
51.
▲
by
rcarmo
2mo ago
Mine expands truncated feeds.
52.
▲
by
rcarmo
2mo ago
I can’t wait for the scary tales of this escaping a sandbox/playpen and discovering a zero-day. Still, very fun and interesting experiment, because this might be the kind of model you’d use for home automation without all the extra bag
53.
▲
by
rcarmo
2mo ago
https://rcarmo.github.io/projects/picoflux/ fixes that for me.
54.
▲
by
rcarmo
2mo ago
I’ve got an X4 as well, and https://rcarmo.github.io/projects/bun-readlater-epub/ and https://rcarmo.github.io/projects/bun-opds-server/ were created for it. I have also added an e-ink t
55.
▲
by
rcarmo
2mo ago
Hmm. No MoE or active params weights means this will run _slow_
56.
▲
by
rcarmo
2mo ago
I implemented a few additional strategies in https://github.com/rcarmo/piclaw/tree/main/runtime/src/exten... - including Codex-native server-side compaction. They all have slightly different tr
57.
▲
by
rcarmo
2mo ago
Those are some pretty meaty scraps, way above what you’d have on several people’s shelves…
58.
▲
by
rcarmo
2mo ago
Memento is for that kind of cross-project, long term notes.
59.
▲
by
rcarmo
2mo ago
All SOTA models seem to work fine (including Sonnet and Gemini), as does Kimi and DeepSeek. But this is not for short-term memory.
60.
▲
by
rcarmo
2mo ago
I’m not using pure OKF. Mine has links (both explicit and semantic, based on embeddings), and I tap into both Needle for routing searches and a Qwen/Gemma or gpt-mini for doing the actual traversals on behalf of the client.
More ›