Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
jatins
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
12 ms
·
211.
▲
by
jatins
2y ago
If I am reading the table correctly they are claiming it is better than all models but 3.5-Sonnet Is anyone with X premium able to confirm the vibe check -- Is the model actually good or another case of training on benchmarks?
212.
▲
by
jatins
2y ago
Extremely fluffy piece. 20% in and not one valuable piece of information
213.
▲
by
jatins
2y ago
don't think 2 months is generous
214.
▲
by
jatins
2y ago
Well written and relatable. I think Twitter/X has done and continues to do a lot of damage to startup ecosystem. You often meet non-technical folks slinging around buzz words like "let's think from first principles" or &
215.
▲
by
jatins
2y ago
You are seriously overestimating the engineering practises at these companies. I have worked in "enterprise security" previously though not at this scale. In a previous life I worked with of the engineering leaders currently at Cr
216.
▲
by
jatins
2y ago
If you want to make a case against SPA/Angular just show someone the monstrosity GCP dashboard is. Visually ugly, poor performance, functionally buggy with no sign of support for an independent developer. I know all of that is not Angu
217.
▲
by
jatins
2y ago
this was the same argument that was presented a decade ago on why Google was supposed to win the cloud because their internal infra was miles ahead of Amazon and Microsoft. Yet here we are. Will the consumer video cards get cheaper and bett
218.
▲
by
jatins
2y ago
In the current LLM tech if you need sources reliably, you need to start with search first and then summarise that with LLMs. ChatGPT without web browsing skips the search part and ends up hallucinating URLs
219.
▲
by
jatins
2y ago
is Fleet still being developed? I had a feeling it was dead given that Jetbrains has been pretty mum about it since the announcement couple years ago
220.
▲
by
jatins
2y ago
"web frameworks are excellent for web development" yes, that is correct.
221.
▲
by
jatins
2y ago
this is incorrect. Private company RSUs often have double trigger with second trigger being IPO/exit. The "semi" vested RSUs can expire if the company does not IPO in 7 years.
222.
▲
by
jatins
2y ago
couple red flags: 1/ fake testimonials 2/ license that says "in case of conflict the terms of appendix will take precedence" followed by an appendix which says "yeah please don't change my code" seems lik
223.
▲
by
jatins
2y ago
IMO what has ruined React is the emergence of meta frameworks that are tied to vested interests of corresponding hosting platforms. React docs themselves recommend starting with Next, Remix, or Gatsby. I can't imagine how Next or Remix
224.
▲
by
jatins
3y ago
I think the HN crowd loves the attack on management consultant types. However, this reads like an over exaggerated Fox news story, trying to create a hero vs villain narrative, written solely for the purpose of personal attack on a person
225.
▲
by
jatins
3y ago
Does Elixir ML ecosystem libs (Nx, Axon) provide some sort of interop with Python ecosystem? For example can I load or fine tune a model pre-trained in pytorch/JAX in Axon? Or does everything need to be written from ground up in Elixir
226.
▲
by
jatins
3y ago
Clouflare has a lot of great stuff for hosting and databases but I think they haven't done a great job marketing themselves as developer platform which has lead to platforms like Vercel, Netlify taking significant mindshare. Tangential
227.
▲
by
jatins
3y ago
What do you use it for i.e what kind of codebases are you typically working with?
228.
▲
by
jatins
3y ago
> Ban everything you cannot control / understand! That is the classic Indian bureaucracy reaction Reminds me of how India's leading research institute removed ceiling fans to prevent student suicides https://www.indi
229.
▲
by
jatins
3y ago
I read that post recently and it felt prescient to someone who has not been deeply involved in ML Even the HN discussion around this had comments like "this feels my baby learning to speak.." which are the same comparisons people
230.
▲
by
jatins
3y ago
Yeah, you are right on the point that if you are not able to exercise now you might not be able exercise later also. It's still lesser of the 2 evils imo because: 1/ many companies have liquidity events as they mature. So you may
231.
▲
by
jatins
3y ago
Story of bad management aside, this was also a highlight for me on the risks of early stage startup even when the startup goes big > In my case the amount of taxes would be so high I wouldn't be able to afford it, forcing me to wait
232.
▲
by
jatins
3y ago
> - slightly worse at coding > - Worse at logic (e.g. it contradicts itself in a single sentence, and is unable to figure it out) That takes most of my use cases. "logic" is what makes GPT often feel like AGI. Use cases like
233.
▲
by
jatins
3y ago
Does Amazon have no review/verification for what gets uploaded on their store?
234.
▲
by
jatins
3y ago
oh wow, I thought this was new. Given that this has received no attention in 3 years I'd assume this is largely abandoned internally inside company.
235.
▲
by
jatins
3y ago
How small are these models? can I export a model here and embed it in an Android/ios app?
236.
▲
by
jatins
3y ago
Can a person just go upload anything on arxiv or is there a review process around these things? What I am really asking is "what makes something a paper and not a blogpost"?
237.
▲
by
jatins
3y ago
> I'd be interested to hear these horror stories about Msft I believe it's called Ohome and is an ancient tool still used in Windows/office org
238.
▲
by
jatins
3y ago
Which BigTech (10k+ employees) has the worst internal dev tooling? I have heard horror stories about some build systems at Microsoft but curious how others fare.
239.
▲
by
jatins
3y ago
> He was even trying to depose board members who were openly critical of open AI's practices. Was there any concrete criticism in the paper that was written by that board member? (Genuinely asking, not a leading question)
240.
▲
by
jatins
3y ago
> Furthermore, the overwhelming groupthink shows there's clearly little critical thinking amongst OpenAI's employees either. If the "other side" (board) had put up a SINGLE convincing argument on why Sam had to go may
More ›