haritha1313

5 ms·

1.
The Foothills of Bay Area House Party – By Scott Alexander(astralcodexten.com)
17 points by haritha1313 28d ago|1 comments
2.
by haritha1313 1mo ago
users have complete control of what a good output looks like. we have seen that adding criteria that is specific to what the expected output is, including at trajectory level, helps a lot to avoid false positives
3.
Show HN: Oqoqo – build evals and custom benchmarks for real-world tasks(oqoqo.ai)
1 points by haritha1313 1mo ago|2 comments
4.
Unpacking ChatGPT Work: The Agent for a Billion Users(latent.space)
3 points by haritha1313 1mo ago|0 comments
5.
The Next Five Years – Economics of AI – Patrick Collison(next-five-years.vercel.app)
4 points by haritha1313 1mo ago|0 comments
6.
by haritha1313 2mo ago
This just seems like lousy testing. Why was the guardrail just an understanding with the third party and a prompt and not actually tested for edge cases? Starting to wonder if its because of agents monitoring agents' work and yolo-ing
7.
From GPT2 to Kimi3(twitter.com)
2 points by haritha1313 2mo ago|0 comments
8.
Understanding Is the New Bottleneck(geoffreylitt.com)
3 points by haritha1313 2mo ago|0 comments
9.
by haritha1313 2mo ago
?
10.
The Joy of Being Wrong(shortdiv.com)
3 points by haritha1313 2mo ago|0 comments
11.
by haritha1313 2mo ago
openai just signed the letter
12.
Open Weights and American AI Leadership(microsoft.com)
13 points by haritha1313 2mo ago|6 comments
13.
BackSearch – Letting agents search the web(gr.inc)
2 points by haritha1313 2mo ago|0 comments
14.
AI's most important protocol is getting a little bit easier to use(techcrunch.com)
3 points by haritha1313 2mo ago|0 comments
15.
The Joy of Being Wrong(shortdiv.com)
2 points by haritha1313 2mo ago|0 comments
16.
Vibe Maintainer by Steve Yegge(steve-yegge.medium.com)
2 points by haritha1313 2mo ago|0 comments
17.
Chasing the AI High: Clay, Kilns, and the Red Queen's Race(blog.moot.dev)
1 points by haritha1313 2mo ago|0 comments
18.
by haritha1313 2mo ago
I keep buying the pretty notebooks and collecting them. I hate writing on them because they are too beautiful. I've finally started using random company merch notebooks (that are not pretty) while I also carry around a pretty one for t
19.
The human-in-the-loop is tired(pydantic.dev)
318 points by haritha1313 2mo ago|200 comments
20.
Cerebras Built Its Enterprise Knowledge Base(cerebras.ai)
2 points by haritha1313 2mo ago|0 comments
21.
Linux kernel won't be anti-AI – Linus Torvalds(lore.kernel.org)
11 points by haritha1313 2mo ago|4 comments
22.
Harness Engineering for Self-Improvement – Lil'Log(lilianweng.github.io)
2 points by haritha1313 2mo ago|0 comments
23.
Measuring Agents in Production – ICML(arxiv.org)
2 points by haritha1313 2mo ago|0 comments
24.
by haritha1313 2mo ago
The reality is most people building their own models and providing that alongside SOTA ones don't really care about how great these models are. They just prove that 'hey we are smart enough to build our own models so you can trust
25.
Microsoft Replaces OpenAI, Anthropic with Own AI in Some Apps(finance.yahoo.com)
2 points by haritha1313 2mo ago|0 comments
26.
Nvidia CEO blasts bosses using AI as a layoff excuse: 'We're scaring people'(finance.yahoo.com)
4 points by haritha1313 2mo ago|0 comments
27.
Gemini's personalized AI image generation is now free for US users(techcrunch.com)
3 points by haritha1313 3mo ago|0 comments
28.
Improved LLM as a Judge Techniques(arxiv.org)
2 points by haritha1313 3mo ago|0 comments
29.
Data Benchmarks and Limitations [video](youtube.com)
2 points by haritha1313 3mo ago|0 comments
30.
by haritha1313 3mo ago
Ah this is so cool! Wish I had it in my research days.

More ›