Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
smy20011
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
6 ms
·
31.
▲
by
smy20011
11mo ago
Just don't play the game have winner and loser. Play the game that both side can win.
32.
▲
The lawyer who beat Tesla is ready for 'round two'
(theverge.com)
22 points
by
smy20011
1y ago
|
2 comments
33.
▲
Running Jujutsu with Claude Code Hooks
(matthewsanabria.dev)
3 points
by
smy20011
1y ago
|
0 comments
34.
▲
by
smy20011
1y ago
> Often, it will drop you precisely at that golden moment where shit almost works, and development means tweaking code and immediately seeing things work better. That dopamine hit is why I code. Only if you are familiar with the project&
35.
▲
by
smy20011
2y ago
Yeah, fully local LLM+TTS setup. Use jupyter notebook to fetch the stock and weather info and feed that into a local LLM and convert that to speech using opensource TTS. https://github.com/smy20011/MorningRadio
36.
▲
by
smy20011
2y ago
Local server. https://github.com/smy20011/MorningRadio
37.
▲
Xiaomi Su7 has surpassed Tesla's Model 3 in China
(electrek.co)
7 points
by
smy20011
2y ago
|
0 comments
38.
▲
by
smy20011
2y ago
I use tailscale to build my personal podcast that include local weather and stocks I interested in. Running the whole pipeline on a steamdeck and use tailscale to securely delivery the generated podcast to my phone.
39.
▲
by
smy20011
2y ago
Curious why large open source project (like linux) have no manager for all the political stuff but works well but every private company have a huge management chain for less complex stuff.
40.
▲
by
smy20011
2y ago
Not that simple, It could cause resource curse [1] for developers. Why optimize algorithm when you have nearly infinity resources? For deepseek, their constrains is one of the reason they achieve breakthrough. One of their contribution, fp8
41.
▲
by
smy20011
2y ago
Human do write code that scalable with compute. The performance is always raw performance * software efficiency. You can use shitty software and waste all these FLOPs.
42.
▲
by
smy20011
2y ago
Did they? Deepseek spent about 17 months achieving SOTA results with a significantly smaller budget. While xAI's model isn't a substantial leap beyond Deepseek R1, it utilizes 100 times more compute. If you had $3 billion, xAI wo
43.
▲
by
smy20011
2y ago
Looks like the beginning of the downfall of America.
44.
▲
Measles outbreak erupts in one of Texas' least vaccinated counties
(arstechnica.com)
11 points
by
smy20011
2y ago
|
8 comments
45.
▲
Morning Radio – Locally Generated Personal Morning Broadcast
(github.com)
1 points
by
smy20011
2y ago
|
0 comments
46.
▲
Edge TTS
(github.com)
107 points
by
smy20011
2y ago
|
65 comments
47.
▲
by
smy20011
2y ago
Do you have an example of such account? Most of the thing I saw is engagement bait.
48.
▲
by
smy20011
2y ago
It seems O3 following trend of Chess engine that you can cut your search depth depends on state. It's good for games with clear signal of success (Win/Lose for Chess, tests for programming). One of the blocker for AGI is we don&#x
49.
▲
Are We Full Self Drive Yet?
(smy20011.github.io)
7 points
by
smy20011
2y ago
|
2 comments
50.
▲
Waymo – Avoiding a Falling Skateboarder
(twitter.com)
15 points
by
smy20011
2y ago
|
13 comments
51.
▲
The Illusion Illusion
(bsky.app)
3 points
by
smy20011
2y ago
|
0 comments
52.
▲
by
smy20011
2y ago
Mind share the list? Asking for a friend.
53.
▲
Bluesky WebSocket Firehose
(simonwillison.net)
3 points
by
smy20011
2y ago
|
0 comments
54.
▲
by
smy20011
2y ago
Then why BlueSky is #1 free apps while twitter is not?
55.
▲
In-Context LoRA for Diffusion Transformers
(ali-vilab.github.io)
1 points
by
smy20011
2y ago
|
0 comments
56.
▲
by
smy20011
2y ago
The top developer is the one who committed five PRs that changed Godot to Redot, but they had fewer than ten PRs the last time I checked.
57.
▲
Top redot developers left redot and announced a new fork of Godot
(twitter.com)
7 points
by
smy20011
2y ago
|
4 comments
58.
▲
by
smy20011
2y ago
A pretty good UX improvement for ChatGPT. I think they will fork VScode and make an AI editor later. Although It's confusing for me the OAI seems spend more time on end user product instead of working on AGI.
59.
▲
by
smy20011
2y ago
Numbers are impressive, not sure why it's down in AH.
60.
▲
Mistral Remove "Committing to open models" from their website
(old.reddit.com)
197 points
by
smy20011
3y ago
|
56 comments
More ›