prats226

6 ms·

1.
Ask HN: If you are an agent builder, what memory platform are you using?
1 points by prats226 6d ago|0 comments
2.
Ask HN: Best PR Review Experience?
3 points by prats226 3mo ago|1 comments
3.
by prats226 5mo ago
A good experiment would be to also try giving it access to latency traces so it can identify issues? Wrt coding agents, giving access to observability tools often improve coding/debugging ability for me
4.
by prats226 7mo ago
Try https://docstrange.nanonets.com/ once, 10k docs you can use for free. Strong table performance. Do give feedback if any. Powered by bigger model compared to our open source one which is quiet popular on HF.
5.
by prats226 7mo ago
If with LLM's you can deanonymize at scale, on a personal level, you should also be able to figure out what posts are leading to this deanonymization and remove them or modify them.
6.
by prats226 8mo ago
Instead of markdown -> LLM to get JSON, you can just train a slightly bigger model which you can constrain decode to give JSON rightaway. https://huggingface.co/nanonets/Nanonets-OCR2-3B We recently published a cook
7.
by prats226 8mo ago
https://nanonets.com/cookbooks/structured-llm-outputs/uncons...
8.
by prats226 8mo ago
Nice, it would be good idea to develop CFG for this as well so can embed it into all these constrained decoding libraries
9.
by prats226 8mo ago
One of the authors here, will checkout the diagram link. Every commercial model provider is adding structured outputs so will keep updating the guide.
10.
by prats226 11mo ago
https://docstrange.nanonets.com/ as well, wrapper on top of 7B version of https://huggingface.co/nanonets/Nanonets-OCR2-3B
11.
by prats226 11mo ago
Then you can just download finetuned version of same multi-modal foundation model that's trained on documents?
12.
by prats226 11mo ago
Top 3 models on huggingface are all OCR models. Most automation projects involve documents where you need a model finetuned to understand all elements inside documents and provide grounding and confidence scores etc which is why these subse
13.
by prats226 11mo ago
Would be intersting to see where funding goes to fix these issues. News would heavily impact public opinion and hence political influence and public funding.
14.
by prats226 11mo ago
Yes, and its not just OCR (Optical Character Recognition), it understands layouts, captures signatures, charts, watermarks etc so way beyond just characters
15.
by prats226 11mo ago
https://mention.com/en/
16.
by prats226 11mo ago
Here is link to open source model: https://huggingface.co/nanonets/Nanonets-OCR-s And hosted model: https://docstrange.nanonets.com/
17.
by prats226 1y ago
It boils down to information loss in compaction driven by LLM's. Either you could carefully design tools that only give compacted output with high information density so models have to auto-compact or organize information only once in
18.
by prats226 1y ago
Reason I felt like they are closely connected are because for designing tools for lets say coding agents, you have to be thoughful of context engineering. Eg linear MCP is notorious for giving large JSONs which quickly fill up context and h
19.
by prats226 1y ago
Context engineering is another name people have given to same skill?
20.
by prats226 1y ago
You can always put automation for your google home to blast music at full volume at right time. And if you don't wake up from sound of music yourself, your neighbour will knock on your door for sure!
21.
by prats226 1y ago
With google serving AI overviews, now an average search query should cost more? Compute is getting cheaper but also algorithms getting more and more complex, increasing compute?
22.
by prats226 1y ago
Read long time ago that even SFT for conversations vs base model for autocomplete reduces intelligence, increases perplexity
23.
by prats226 1y ago
> But here’s the important part: LLMs don’t know how to use tools. They don’t have native tool calling support. They just generate text that represents a function call. Its not a completely true statement. Eg openAI uses libraries like l
24.
by prats226 1y ago
This is super interesting to think about in LLM world where lot of software is getting replaced with LLM calls. In terms of output of an LLM, there is no clear promise in the contract, only observable behaviour. Also the observable behaviou
25.
Ask HN: Is OpenAI charging me for free tokens?
1 points by prats226 1y ago|1 comments
26.
by prats226 1y ago
Can you give a specific example where at certain depth it has stopped becoming useful?
27.
by prats226 1y ago
I see that in tool calling, we usually specify just the inputs to functions and not what typed output is expected from function. In DSL style agents, giving LLMs info about what structured inputs are needed to call functions as well as what
28.
by prats226 1y ago
https://www.jonathansapps.com/ Someone recently suggesting this and first thought I came to my mind is prevailing advice is to not show kids screen as long as possible
29.
by prats226 1y ago
Interestingly, deepseek paper mentions RL with process reward model. However they mentioned it failed to align model correctly due to subjectivity involved in defining if the intermediate step in process is right or wrong
30.
by prats226 1y ago
Isn't it because there is a difference in your field and other fields? 1) Scope - Other fields like law, medicine atleast are impacting one unit at a time, vs software which is impacting large number of users through your work. I am su

More ›