Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
witnessme
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
5 ms
·
31.
▲
by
witnessme
1y ago
How fast can you react to the learning you have from the market, that's the bottleneck. And yes, the development is the big chunk of that reaction time.
32.
▲
by
witnessme
1y ago
I like the DSL approach but can't imagine how practical and effective it is. Specially considering the cost.
33.
▲
by
witnessme
1y ago
This latest test is related to rooting framework KernelSU. Sinilar vulnerabilities were found earlier in other rooting frameworks such as APatch , SKRoot, etc. The vulnerability results in gaining root access and escalated privileges.
34.
▲
Rooted Android phones vulnerable due to Android kernel patching flaws
(zimperium.com)
18 points
by
witnessme
1y ago
|
11 comments
35.
▲
A Single Poisoned Document Could Leak 'Secret' Data via ChatGPT
(wired.com)
8 points
by
witnessme
1y ago
|
0 comments
36.
▲
by
witnessme
1y ago
I see you use the Zoekt project for code search. Why did you choose this over alternatives and how has been your experience so far?
37.
▲
by
witnessme
1y ago
Thanks for making it clear
38.
▲
by
witnessme
1y ago
Surprised to find out grok 3 mini is so economic and ranks higher than equivalent gpt models. I run most of my agents on gpt4.1 mini, might switch now
39.
▲
by
witnessme
1y ago
We are in the same group then
40.
▲
by
witnessme
1y ago
What is the infrastructure tooling behind this product? I wanted to build this but it seemed like it is not a small startup project to build such big infra without riding on the existing email infra (which results in loss of the excitement
41.
▲
by
witnessme
1y ago
Snippet from the repo > Thorium is a scalable file analysis and data generation platform that allows users to easily orchestrate arbitrary docker/vm/shell tools at scale. Features: * Highly scalable analysis of arbitrary files&
42.
▲
Open Source malware and forensic analysis tool by CISA – Thorium
(github.com)
4 points
by
witnessme
1y ago
|
3 comments
43.
▲
Ask HN: Tell me your horror story about blowing money on AI
7 points
by
witnessme
1y ago
|
3 comments
44.
▲
by
witnessme
1y ago
Did you try it? How was your experience?
45.
▲
by
witnessme
1y ago
Before deployment: Use tools such as codeql, sonarqube, etc. in your CI/CD pipeline. They scan the repo for known patterns and you can design your own patterns to automatically check before anything goes out. After deployment: There wa
46.
▲
by
witnessme
1y ago
Love the UX. From the value POV, I am yet to see/experience how it differs from competitors. P.S. I currently use Braintrust and Opik
47.
▲
by
witnessme
1y ago
Well, they gotta do what they gotta do. But as a developer, this kills the positioning and trust it had for me. I do not see it as a developer tool project anymore.
48.
▲
Gemini CLI vulnerability allows hackers to execute hidden malicious commands
(arstechnica.com)
8 points
by
witnessme
1y ago
|
3 comments
49.
▲
Efficient algorithm by MIT solves AI symmetry problem impacting drug discovery
(news.mit.edu)
1 points
by
witnessme
1y ago
|
0 comments
50.
▲
by
witnessme
1y ago
Is it? I thought this is a genuine article
51.
▲
by
witnessme
1y ago
Omegle is hard to kill. Is the code Open Source?
52.
▲
by
witnessme
1y ago
How different it is from node-red?
53.
▲
by
witnessme
1y ago
My common problem with all these visual builders is that they get too complex to maintain, all my workflows built with them are sitting there for years, I fear to touch them again. On the contrary the benefit these visual builders offer is
54.
▲
GPS vs. BDS- Iran banned GPS and switched to Chinese BDS
(lidarandradar.com)
2 points
by
witnessme
1y ago
|
2 comments
55.
▲
by
witnessme
1y ago
Interesting idea. I was expecting a page to explore videos directly but it shows me the text box to paste link and get summary. The same could be done via chatgpt, I may try if I get to experience how my youtube search/explore is chang
56.
▲
by
witnessme
1y ago
HRM (hierarchical reasoning model) does make sense (breaking down thinking to hierarchical reasoning - a loop of high level planning + low level computation). Does it work? The benchmark shown are impressive. Having said that, I doubt if it
57.
▲
by
witnessme
1y ago
Couldn't agree more