Viewing profile — ddematheu

Interesting performance with GPT 3.5, what does performance on Llama look like? What about smaller models like Llama 3.1?

I don't disagree with all your points. That said, what we have built has proven useful for us as we have built pipelines for customers and think it might be useful for others. Prob…

LlamaIndex is pretty awesome. There are a couple areas where we think we are driving some differentiation. 1. The management of metadata as a first class citizen. This includes cap…

It is dangerous, part of the reason that we haven't productized that further. One of the ideas we had to productize the capabilities further was to leverage edge / lambda functions…

Yeah, we were playing around with doing some semantic chunking. Works okay for some use cases. We have some ideas to go further on that. Generally we have found that recursive chun…

Haven't connected.

Co-founder here :) Today, it is mostly about convenience. We provide abstractions in the form of a pipeline that encompasses a data source, embed and sink definition. This means th…

Lies or not lies, the point was the train on the authentic message that the candidate wanted to provide. Try to be as unbiased as possible.

How real-time is it? Just app or API?

Disclaimer: I am the co-founder of Neum AI, a data processing platform for vector embedding generation and data retrieval for RAG. With the upcoming elections 2024 coming up, we wa…

Some engineers find it fun, other might not. Same as everything. IMO the fun parts are actually prototyping and figuring out the right pattern I want to use for my solution. Once y…

What about then sucked?

Co-author of the article here. We do support updates for some sources. Deletes not yet. For some sources we do polling which is then dumped on the queues. For other we have listene…

Through the platform (Neum AI) we support the ability to do this with Postgres, it is just a cloud platform so not a python library. Curious on what type of customization are you l…

To some degree. The amount of data that will be brought into search solutions will be enormous, seems like a good time to try to reimagine what that process might look like

Makes sense. Interesting on the fact that summaries affect quality sometimes. For synthetic data scenarios are you also thinking about synthetic queries over the data? (Try to pred…

The queues and storage are the foundation on which some of these other integrations can be built on top. Agree fully on the need for LLMs within the pipelines to help with data ana…

What type of latency requirements are you dealing with? (i.e. look up time, ingestion time) Were you using postgres already or migrated data into it?

Not at scale. Currently we do some extraction for metadata, but pretty simple. Doing LLM based pre-processing of each chunk like this can be quite expensive especially with billion…

Co-author of article here. Yeah a ton of the time and effort has gone into building robustness and observability into the process. When dealing with millions of files, a failure ha…

Co-author of the article here. You are right. Retrieval accuracy is important as well. From an accuracy perspective, any tools you have found useful in helping validate retrieval a…

TL;DR for blog: If you are using RAG across structured data, make sure you consider the role of metadata as it is likely that not all the fields within your data carry semantic mea…

Q&A with documents is the most common use cases for LLMs today. But scaling from one or two documents to thousands can be challenging. This blog explores the challenges in scaling …