zurfer

25 ms·

181.
by zurfer 2y ago
Anthropic recently released something that looks more polished but follows the chat paradigm. [1] As a builder of something like that [2], I believe the future is a mix, where you have chat (because it's easy to go deep and refine) AND
182.
by zurfer 2y ago
There are a bunch that I put into this list: https://github.com/Snowboard-Software/awesome-ai-analytics ( probably missing a similar amount since many people try) I would be curious to understand why you are looking fo
183.
by zurfer 2y ago
Here is a link to the prompt. https://github.com/anthropics/anthropic-quickstarts/blob/9d5... On a second glance I'd say that the visualizations are pretty limited but good enough for a demo.
184.
by zurfer 2y ago
Well done. We are building something similar [1] and found that generating UIs (mostly charts) on the fly works surprisingly well in most cases, but can be a bit frustrating if you know exactly what you want and just can't prompt it (a
185.
by zurfer 2y ago
too late to edit now. I was completely wrong about open-weights. The meme at the bottom made me jump to that conclusion. Well, not that exciting of a release then. :(
186.
by zurfer 2y ago
poor title, Mistral released new open weight models that win across benchmarks in their weight class: Ministral 3B and Ministral 8B
187.
by zurfer 2y ago
I don't know what YC thinks, but the mission is not about prestige, but to help startups succeed, so these startups can make something people want. Also, YC is mostly back in person and so pretty hard to scale (also batch sizes have go
188.
by zurfer 2y ago
Kudos to the author who is responsible for the whole stack. A lot of effort goes into ingesting data into Iceberg tables to be queried via AWS Athena. But I think it's great that analytics and data transformation is distributed, so dev
189.
by zurfer 2y ago
Amazon was not profitable for many years, yet they are one of the most valuable companies in the world. Why? Because they kept growing. What OpenAI is doing is "sustainable" if they keep growing like that. From summer 2023 to sum
190.
by zurfer 2y ago
I understand how (a) discoverability and (c) recoverability are a problem, but what do you mean with (b) efficiency? Most people talk faster than they can type.
191.
Dot (YC S21) Is Hiring an Engineer to Automate Analytics (Remote)(ycombinator.com)
1 points by zurfer 2y ago
192.
by zurfer 2y ago
OpenAPI is great, we use it in combination with redoc ( https://github.com/Redocly/redoc ) to have a almost 0 effort API for our product. But man, my tokenizer is in trouble with OpenAPI and OpenAI.
193.
by zurfer 2y ago
It is already said that gpt4 was trained on all high quality internet data. So it should have been included already. It seems to me that o1 has the same/similar pretraining corpus. So we have 3 options: - t3 was now included in the cor
194.
by zurfer 2y ago
sorry for that. I just learned that my team account does not allow me to share conversations publicly. Pasting them here for reference * Conversation 1 let's play tic tac toe. you need to play optimally ChatGPT Sure, I&#
195.
by zurfer 2y ago
> I asked on Twitter for examples of prompts that people had found which failed on GPT-4o but worked on o1-preview. it seems trivial, but I tried for more than 2 hours in the past to get gpt4 to play tic-tac-toe optimally and failed (CoT
196.
by zurfer 2y ago
even easier, just a job board (like stackoverflow), where companies pay to advertise a job.
197.
by zurfer 2y ago
I love that you can take a screenshot and it starts to explain what it sees! While this is clearly an ai analytics assistant your "retrofit" approach certainly differentiates you from existing approaches: https://github
198.
by zurfer 2y ago
Could this be part of your competitive advantage in the future? or is the software a commodity and just kind of annoying at times? Changing ERP systems takes a long time (months to years) and for the whole transition period you have TWO sys
199.
by zurfer 2y ago
I want that in a video game, not in customer support.
200.
by zurfer 2y ago
Ok I'll bite. Imitation intelligence is a good term to describe the idea that LLMs "just" auto complete based on their training data. And there is truth to that. If the task you try to do was in pre training, you can expect
201.
01.ai Yi-Large LLM Launch(01-ai.github.io)
1 points by zurfer 2y ago|1 comments
202.
by zurfer 2y ago
Fireworks makes that available with ~200 t/s (claims 300t/s). Ranks higher than llama3 on lmsys (right after GPT-4-0125-preview)
203.
by zurfer 2y ago
While this is interesting, I wonder what the use case is, other than better autocomplete?
204.
by zurfer 2y ago
Congrats on the launch and kudos for open sourcing everything. I'll put you on this list: https://github.com/Snowboard-Software/awesome-ai-analytics One thing I'd love more clarity on as a user would be how n
205.
by zurfer 2y ago
The certainty in this sentence is not appropriate. You refer to an evolutionary process that is highly random and inefficient. We don't have millions of tries to get this right.
206.
Hacker News with headline-based topic extraction(hn-with-topics.web.app)
2 points by zurfer 2y ago|2 comments
207.
by zurfer 2y ago
Last time I heard of Rockset was at the Snowflake Summit where they positioned as a faster DWH. Looking at the landing page now it seems they almost pivoted into semi/unstructed data. To your point, I feel like nobody knows exactly how
208.
by zurfer 2y ago
This is interesting and I think it's smart to focus on marketing integrations first with out of the box connectors. I'll put you on this list: https://github.com/Snowboard-Software/awesome-ai-analytics One th
209.
by zurfer 2y ago
This is impressive. I've just ran a couple of gpt4o workloads for getdot.ai on Sonnet and the quality is great. Plus it's super fast right now ~110 token/s (but gpt4o was also super fast when they launched). But what will sta
210.
by zurfer 2y ago
yeah it's off somehow. rule 1: start at the green dot? rule 2: glue the left outer piece to the bottom rule 3: overlap every now and then :D rule 4: invert some of the pieces every now and then

More ›