Yevgen Reztsov

The DataFusion SIGMOD paper is officially out. Congrats to Andrew Lamb and the other authors! The paper is great, accessible read covering the architecture of a modern, state-of-the-art query engine. Also exciting to see Arroyo (YC W23) get a mention! The coolest thing about the paper (and about DataFusion itself) is that it demonstrates how building the engine in an open, extensible way doesn't preclude great performance. Systems built on top of DF (like Arroyo) can easily add new functions, operators, tree-rewrites, optimizations, data types, and more while using the the same tools and infrastructure that the built-in functions/operators/etc. are built with. This is incredibly powerful, and has already led to a blossoming of powerful data systems including Comet (a Spark accelerator), LanceDB (a vector DB and query engine), and ParadeDB (which embeds an OLAP DB in postgres). Read the paper here: https://lnkd.in/dpknuPy4

33

2 Comments

The game of large language models (LLMs) got a major plot twist! Meta's unveiling of Llama, a family of AI models pushing the boundaries of parameter size and processing power, is a game-changer. Why it matters? Llama boasts a staggering 405 billion parameter model, dwarfing previous heavyweights like Google's LaMDA and OpenAI's Jurassic-1 Jumbo. This sheer size unlocks unprecedented capabilities in: -> From scientific papers to philosophical treatises, Llama can digest vast amounts of data, leading to deeper comprehension. -> Imagine writing styles that perfectly mimic different authors or create compelling narratives on the fly. -> From machine translation to chatbot interactions, Llama holds the potential to transform how humans and machines communicate. -> Llama isn't just about brawn. Meta stresses its commitment to open-source access and responsible AI development. This fosters collaboration and helps mitigate potential biases often associated with large datasets. Meta's ambitious project is a harbinger of the future, where AI can not only process information but truly understand it. Imagine personalized education tailored to individual learning styles, or AI assistants that anticipate your needs before you even ask. https://lnkd.in/dYn_cTjB #MetaLlama #ai #LLMs #futureoftechnology #naturallanguageprocessing #opensource

6

Large language models are wonderful at ingesting large amounts of content & summarizing. Uploading an academic paper and I can pester it with an infinite list of questions & it will respond with equally infinite patience. In comparing the two most recent Microsoft earnings calls, Claude highlighted: - faster than expected Azure growth (29% vs 27%) - AI contributing 8% of Azure revenue up from 7% - higher CapEx spending & greater capacity constraints for data centers - much better commercial bookings growth : 17% vs 13% Excellent analysis. Benn Stancil described LLMs as great averagers of information. But what if I don’t want the average? If I seek the data point two standard deviations out? If I’m channeling my inner Anthony Bourdain & I’m seeking the fermented shark or cobra heart within my query? At Google, we ranked web pages & ads with many signals. An engineer taught me that of them found rarity of a word across a set of documents. For example, if there’s a document containing 10 instances of the word ghotiacross a collection of documents, where the average document has 0 ghotis, that document is likely the best to answer a search about fish. I haven’t found a way to goad an LLM to produce the rare result. Why did one user decide to use a particular piece of software when 10 others did not? Maybe I haven’t learned how to prompt an LLM well. Ideogram launched a feature called Magic Prompt that expands a basic prompt into richer instructions that elicit better results, marrying the language of a user & a computer. A user might write : A boy with a dog in a park. Magic Prompt replies : A heartwarming scene of a young boy playing with his happy, wagging dog in a lush, green park. The boy has a playful smile on his face while he tosses a bright yellow ball for his furry friend to fetch. The park is filled with beautiful flowers and trees, creating a serene atmosphere for this bonding moment between boy and dog. Magic Prompt surprised me in three ways. First, unlike Google search where terse queries tend to work very well, LLMs relish verbosity. Less Hemingway & more Faulkner. Second, maybe there’s a prompt for rarity. If you know what it is, please share with the rarity shibboleth. Third, I wonder what other prompt techniques are out there that have yet to be discovered to wrest the best from an LLM. This technique is called TF-IDF. It means Term Frequency Inverse Document Frequency.

20

5 Comments

People rightly state that there is a lot of hype surrounding #AI. This reflects our obsession with applying new tools to *every* problem, fit for purpose or not. 😄 Here's the good part, though: when applied properly, AI is a huge force multiplier. Imagine a lever ^ 1000000. I've been sharing, any time I have the opportunity, the "good parts": areas in which AI can have real utility now. Not someday. Not maybe. Absolutely yes, and now. One area in particular is #RAG, Retrieval Augmented Generation. RAG == Turning an AI loose on YOUR DATA. Imagine all of those documents in which vital information is squirreled away, hard to search, even harder to correlate *across* those myriad documents. What a treasure trove if only you could leverage it! With RAG, you can...and pretty easily, too. Using AI and RAG, 1+1 pretty quickly adds up to 500. Or more. Even now, the power of your AI-enabled application is more limited by your imagination than the capabilities of the tools. If you want to know more, get just a taste of what you can do with AI using RAG, join me this Monday at #RAGHack. See below for more details and to save your spot! https://lnkd.in/gYAsiSP4

13

1 Comment

Super interesting intro to GraphRAG today! https://lnkd.in/gFH8TeBn First John Alexander showed how to use Langchain with Neo4J to ask questions about a movie knowledge graph, then Apurva Mody showed how to use Microsoft's graphrag package to answer questions about a book. Get slides & code from https://lnkd.in/gErJw-gz #graphrag #azure #neo4j #langchain

59

3 Comments

I'm thrilled to see Anthropic's latest announcement of the upgraded Claude 3.5 Sonnet and new Claude 3.5 Haiku models! 🚀 As someone who strategically leverages various AI platforms, including Claude, for specific use cases, these advancements are particularly exciting. The enhanced coding capabilities in 3.5 Sonnet are especially promising 💻✨, as they build on Claude's already impressive performance in this area. What really stands out to me is the groundbreaking computer use capability now in beta. 🤖🖱️ While I currently use AI selectively and thoughtfully for specific tasks, features like this - where AI can interact with computers like humans do - hint at the expanding possibilities ahead. I appreciate Anthropic's transparent approach in acknowledging that this feature is still experimental and may have limitations. This aligns with my philosophy of carefully choosing when and how to implement AI tools as they mature. 🎯 The fact that Claude 3.5 Haiku matches Claude 3 Opus's performance while maintaining cost efficiency is also noteworthy. 📊💪 This kind of advancement helps organizations like ours make informed decisions about scaling our AI usage as the technology evolves. Looking forward to exploring these new capabilities! 🌈 #AnthropicAI #Claude35 #AIInnovation #ComputerUse #FutureOfTech

2

Interesting post by João Salgado about partitioning vs. sharding in #Postgres which lists trade-offs for each. It's great looking at the list of cons for partitioning and seeing that #TimescaleDB actually helps with most of them - Complexity ⬇️ because we transparently manage partition creation and lifecycle for you as data is ingested - Optimization Challenges ⬇️ because we have planner hooks that help with handling data exclusion (you still do need to pay attention to your partitioning key though). - Storage Overhead ⬇️ (although Postgres doesn't really have much overhead for partitions) because you can just choose to compress your data (we often see a 95x reduction in size). Timescale 🚀

18

1 Comment

Messing around with whisper locally, wanted to see how well it worked on some amateur radio recordings ... seems pretty good! Running on an nvidia a2 tensor core GPU. I'll have to try it out on some lower quality recordings, but it would be really cool both for amateur radio and scanner radio applications in near real time. I could see using this for when I'm scanning P25 Phase 2 for local law enforcement and being able to trigger on different kinds of events. Just real time captioning would be great too. Take a listen to some of the amateur radio recordings, they can often be difficult to grok in real time. https://lnkd.in/efpKEsif Anyway, really impressed and will play with some more, check out this (partial) transcription from the 72 rag chew. I wonder what it would take to identify the different operators by call sign... [46:49.480 --> 46:50.700] Why is he up here? [46:52.540 --> 47:00.180] You're, uh, wasn't at your usual, uh, sitting, uh, frequency, so, uh, I, uh, don't know what [47:00.180 --> 47:00.980] net it is. [47:01.460 --> 47:03.720] You said, was there any chance for the net? [47:03.780 --> 47:04.880] And I said, oh, what the hell? [47:04.880 --> 47:11.420] I'll, uh, I'll pop in and say hello, uh, but a good signal. [47:11.660 --> 47:19.060] Uh, nevertheless, as always, you're, uh, gosh, uh, I forget, but you're well over S9, probably [47:19.060 --> 47:22.140] S, uh, probably 20L or something in that neighborhood. [47:22.820 --> 47:30.340] I don't think, uh, I ever have a problem hearing you, so, uh, and I don't know if I'm in the system [47:30.340 --> 47:30.920] or not. [47:30.920 --> 47:39.000] Uh, this is the first time I think I've ever taken this mess, but, uh, you know, uh, I [47:39.000 --> 47:44.400] may have before, and just don't, uh, and just didn't realize it, but like I said, I heard [47:44.400 --> 47:52.840] a, uh, familiar voice, uh, up here, uh, higher in the band, which, uh, caught me by surprise, [47:52.840 --> 47:59.140] so, uh, uh, nice signal in Indiana where it's, uh, 33 degrees.

13

1 Comment

So, (public) Data is no longer the New Oil. What's next for AI? Agents, Agentic systems and (Private) Data are taking over the scene. Take a look at our latest papers for a glimpse into the near future: https://lnkd.in/dFucKH4P https://lnkd.in/dzwXHkFA #artificialintelligence #Conversationalai #generativeai #agenticai #multiagents Open Voice TrustMark Initiative & Open Voice Interoperability Initiative Deborah Dahl Emmett Coin The Linux Foundation

8

1 Comment

#ObsessivelyEnthusiasticInvestor #UnreasonablyEnthusiasticInvestor #IrrationallyEnthusiasticInvestor at REFASHIOND Ventures: The Industrial Transformation Fund. We look for the people we believe are doing the best work in Industrial Transformation & Supply Chain Technology; Attempting the impossible. Then we go all-in on them. With obsessive and unreasonable enthusiasm. Join us. #IndustrialTransformation #SupplyChain #Innovation #Technology #Startups #VentureCapital #DemandTheImpossible #UnreasonableEnthusiasm #ObsessiveEnthusiasm #IrrationalEnthusiasm 

6

And More [BUILD] with the JADC2 SDK: https://lnkd.in/gaBbWRdV Want to your applications to read/write annotations from Gaia? This example React application demonstrates how to use Palantir's public Map Layers API (https://lnkd.in/gdHufpCP) to retrieve an annotation layer from a Gaia map and plot those annotations on a Leaflet map. In this example, we use an open source library, Spatial Illusions, to generate military symbology by SIDC code. GitHub Resource: https://lnkd.in/gBDJKK_h

119

1 Comment

"Which one is faster?" - Great question Introducing: 🤖 LLM Arena 🚥 📊 This app allows you to measure the speed and word count across OpenAI, NVIDIA, and Groq language models API's and compare their performance. Why? This is a solution to a question I had while implementing an agentic solution, I was thinking all the possible ways to make it run faster. I came up with this app just to illustrate the concept of how useful could it be to choose the right models and providers to accelerate your flow. 🛠️ Stack: Langchain, FastAPI, Streamlit and Docker. Take a look at my repo here: https://lnkd.in/e4Q8Q2V5 🎙️This app will be one of many things I'll be covering in my upcoming talk! If you are in Madrid on 📆25th September, see you there! Free 🍕 🍻 powered by InnoIT Consulting and My Tech Plan 👉Register here: https://lnkd.in/ev_-n9vx #genAI #OpenAI #Groq #Nvidia #AI #Langchain

12

3 Comments

Just dropped: A deep dive with GenAI developer and Sam Witteveen on AWS's bold new claims about solving hallucinations in LLMs. AWS just announced they can prevent 100% of hallucinations in certain mission-critical use cases through "Automated Reasoning checks" - a fascinating application of mathematical verification techniques to LLM outputs. How does it work? AWS CEO's Matt Garman explains they're using proven automated reasoning technology (already used to verify IAM policies and S3 storage systems) to mathematically validate model outputs against source documents. [See minute 1:21:00 through 1:25:35 of his keynote: https://lnkd.in/gXWrYvPZ] This can be game-changing if you have ground truth documentation that can be used for the LLM to reference -- for example in the insurance industry that Garman highlights, and other highly regulated industries. On the other hand, this technology shouldn't be seen as a solve for the LLM's fundamental tendency to hallucinate. Sam raises crucial questions: Can you really eliminate ALL hallucinations without compromising an LLM's creative abilities? And what exactly constitutes a "hallucination" in different contexts? We also dig into: • AWS's vision for collaborative AI agents • The real story behind AWS Nova benchmarks • Why Anthropic may be leading the pack on model alignment • Why new offerings from Pydantic (& Samuel Colvin) and Emergence AI signal a shift toward production-ready AI agent tools that can sit on top and/or outside of monolithic tools offered by Microsoft or AWS. • Cohere's innovative multi-model approach to RAG Watch the full discussion: https://lnkd.in/ggBYNERM #ArtificialIntelligence #AWS #LLM #EnterpriseAI #TechNews VentureBeat

45

With the discussions around llms.txt (https://llmstxt.org/) and Model Context protocol or MCP (https://lnkd.in/gDZRGNAa). It's clear we're going to need a new set of interfaces and standards to provide context and capabilities to LLMs and the coming wave of Agents. There's a significant amount or risk and opportunity around discovery and validity of content that is going to have huge ripples as we reconsider how people discover information, products and services. This shift in the interface of value from search/social network to Agent is going to be an exciting space to watch. David Berlind, might be an interesting time to consider an MCP relaunch of programmable web #Agentic #interfaces #API #AI #futurism

2

2 Comments

Read the latest VMblog article, "Open Beats Closed: How #OpenSource Is Changing #DataManagement" by Sida Shen of CelerData. https://lnkd.in/gEUPBFhx #DataAnalytics #DataEngineering #DataLakeAnalytics #DataLake #DataLakehouse

7

"Open beats closed, easy beats hard, and cheap beats expensive." Discover why open source data lakehouses are the future in Sida Shen's latest article. Learn how flexibility, innovation, and transparency are driving the next wave of data management 🔓 https://lnkd.in/gEUPBFhx #DataAnalytics #DataEngineering #DataLakeAnalytics #DataLake #DataLakeHouse

6

🚀 RAG systems can feel complex, but here’s a way to simplify: build your RAG solution using a pipeline as an API. This approach makes it: ✅ Streamlined: The entire process flows smoothly in one organized pipeline. ✅ Transparent: Unlike black-box APIs, you gain full visibility into data flow, understanding each step along the way. ✅ Customizable: Tailor each stage of the pipeline to fit your specific needs. By leveraging a pipeline-based API, you not only make RAG more accessible but also give yourself control over the process. Simplify, see more, and stay adaptable. For more insights, check out George Strong Ph.D post! #AI #LLM #RAG

1