Oras

32 ms·

511.
by Oras 2y ago
That’s the problem right now. Throughout the history, jobs get diminished and new ones came on. However, the transition took many many years to happen. Take an example a conductor in a bus to collect the fare, took many years to replace and
512.
by Oras 2y ago
If anyone is looking to compare results visually, I have created an open source OCR visualiser to help identifying missing elements (especially in tables). https://github.com/orasik/parsevision
513.
by Oras 2y ago
I worked in an e-commerce company before and was responsible for the logistics tech. The first thing to keep in mind, is migration will not be easy, mainly because no one would understand how the current systems work or the decisions behind
514.
by Oras 2y ago
The article is about LLM agents, so I’m asking in context. To clarify more, I see frameworks like CrewAI and similar, with tools even from Microsoft to define these “agents” quickly. But when I tried them, I noticed they are no more than c
515.
by Oras 2y ago
Why can’t this be done with functions? I don’t see why you need the complexity and unpredictability of using “agents” to do that. I might be missing something?
516.
by Oras 2y ago
Genuine question, are there any examples of agents in production?
517.
by Oras 2y ago
I applied for W22, got interviewed and was rejected. The rejection email was quite thoughtful and helpful. I applied twice after that but didn’t get an interview. Fingers crossed for this one. Why I’m applying? 1. Been a daily reader for HN
518.
by Oras 2y ago
Google made many mistakes with AI that their models do not get attention anymore. When Anthropic Sonnet doing great at text and coding, OpenAI with their reach via web, mobile, and desktop app, and free deepseek interface with great coding
519.
by Oras 2y ago
This is great and unbelievably fast! I noticed a small note saying how much this would cost and how many images you can create for $1. I assume you’re offering this as an API? Would be nice to have pricing page as I didn’t see one on your
520.
by Oras 2y ago
Perplexity style is definitely better than Google. I can't remember the last time I googled something. Having it part of the OpenAI subscription, and assuming it will be accurate, I think Perplexity will have a problem. Let's see
521.
by Oras 2y ago
There was a similar app for men's fashion in the UK called Thread, acquired by M&S in 2022. Source: https://corporate.marksandspencer.com/media/press-releases/m...
522.
by Oras 2y ago
This is obviously good news, but __personally__ I feel the open-source models are just trying to catch up with whoever the market leader is, based on some benchmarks. The actual problem is running these models. Very few companies can afford
523.
by Oras 2y ago
Nicely done. I was able to see my tiny contribution to HN clearly. Few questions: 1. Did you create embeddings for the whole comment or you used chunking? 2. How do you determine the effective keywords in comments?
524.
by Oras 2y ago
I always thought booking sites had partnerships with airlines and using an API!
525.
by Oras 2y ago
That depends on how many users are aware of the AI content. HN is not a reflection of the world.
526.
After Apple, Meta also withholding future AI models from EU countries(9to5mac.com)
3 points by Oras 2y ago|1 comments
527.
by Oras 2y ago
Congratulations on launch. This is a crowded market, and there are many tools doing the same thing. How are you differentiating yourself from other tools like: Langfuse Portkey Keywords ai Promptfoo
528.
by Oras 2y ago
That's not about public data. Reference [7] talked about creating fake accounts to scrape. > hiQ had prevailed on the Computer Fraud and Abuse Act (CFAA) “unauthorized access” issue related to public website data but was facing a ru
529.
by Oras 2y ago
I watched some videos about aider today, and some claimed it works well with DeepCoder V2. Haven't tried it yet, but interested to see if you had experience with it?
530.
by Oras 2y ago
Filtering works pretty well, and the design is well executed. For those who complain about scraping, if this is public data, as the OP mentioned, then I don't see how it is different from Google. A clean search would benefit the creato
531.
Show HN: Parse Vision, open source tool to visualise OCR output(github.com)
3 points by Oras 2y ago|0 comments
532.
by Oras 2y ago
Portkey does that with configuration. You assign a base model, then add more models with weight to load-balance.
533.
by Oras 2y ago
Word of caution when using perplexity to summarize URLs. I have tried the API and had lots of hallucinations. It worked better in the UI, I only had it with the API.
534.
by Oras 2y ago
They use a tool with outdated data, and no one checks and validates when they do it at a scale.
535.
by Oras 2y ago
AI or not, cold emailing is dead. I receive tons of these by email and LinkedIn, to the point that I stopped reading them. I talked to many people, and all have developed immunity against the cold outreach.
536.
by Oras 2y ago
The comments are good example that hype > quality. 99% of docs mentioning LangChain or showing a code example with LangChain. Wherever you look at tutorials or YouTube videos, you will see LangChain. They take the credit of being the fir
537.
by Oras 2y ago
Cohere is underrated, I recently tried cohere R model and found it following prompt much better than gpt-4o and even Claude opus. That’s said, it’s a bit annoying to see langchain examples all over. Not everyone uses it, and many consider i
538.
by Oras 2y ago
Can this be dynamic on prompts and providers? I’m thinking of continuous evaluation for LLM in production, where after each call, a webhook will send the input/output to evaluate.
539.
by Oras 2y ago
Would be nice to read how do they collect/prepare data for training. Which data sources? How much of Meta users data (fb, instagram… etc). How do they sanitize PII?
540.
by Oras 2y ago
Your comment suggests people making 1M in FAANG and somehow a stable job. Did you hear about the thousands of layoffs in recent years?

More ›