Luke Kim’s Post

Spice AI is hiring Senior and Principal Rust Engineers in Seattle (we're University of Washington CoMotion Labs). You'll work on high-performance data technologies like Apache Arrow, Apache Iceberg, DuckDB, Apache DataFusion, and the intersection of data and AI for data-driven, agentic RAG. Watch my CMU Databases talk to learn more and apply at https://spice.ai/careers #data #ai #developers #apache #datafusiuon #agents #databases

Hot take. AI will help you collect data and build forms faster and more efficiently. Check it out! #ai #aidatacollection #conversationalai #aiformbuilding #handsfreedatacollection

Super impressed with Snowflake Cortex, now in public preview, surpassing 90% accuracy for real-world use-cases. Another great innovation from Snowflake which will enable more business users to ask questions of their data in Snowflake using natural language to receive reliable answers. Snowflake Cortex is available on the Dataiku #LLMMesh so if you want to experiment with all the great LLM's out there, and build safe, secure and reliable Gen AI applications, check it out!

🚀 Introducing the RAPTOR Workflow: A cutting-edge framework designed to optimize information retrieval and enhance query efficiency. The process begins with Text Chunking, followed by Clustering and Summarization, creating structured and concise data clusters. These clusters are then utilized in Tree Construction to map out the relationships and hierarchies within the data. When a query is processed: The Primary Path involves traversing the tree to perform Similarity Searches, leading to targeted results. The Alternate Path queries all nodes, ensuring a broader search scope. Finally, the results from both paths are merged for comprehensive Information Retrieval, ensuring precision and recall are balanced. Whether you're dealing with vast datasets or complex queries, RAPTOR ensures efficient processing and retrieval, setting a benchmark for intelligent workflows. #GenerativeAI #LLM #ArtificialIntelligence #DataRetrieval #WorkflowOptimization #TechInnovation

Here’s how you can build your own AI Agent, feed it on-chain data, and let it handle your analytics! ⚡️ The ability of AI agents to autonomously handle complex tasks has made them quite popular. That’s why we decided to show you how to make one yourself! 🧑💻 In this video, you’ll learn to: ✔️ Filter the precise data you need for your AI agent using QuickNode Streams ✔️ Store your fetched data in a Postgres database ✔️ Make your AI agent intelligent and responsive by scripting the database ✔️ Bring your AI agent to life on Discord, allowing for real-time interaction and queries ▶️ Here’s the full walkthrough! https://lnkd.in/ec6mqjD6

llmware is a framework for building enterprise grade LLM-based applications (e.g, RAG, Agents), using small, specialized models that can be deployed privately, integrated with enterprise knowledge sources safely and securely, and cost-effectively tuned and adapted for any business process. llmware has two main components: - RAG Pipeline: connecting knowledge sources to generative AI models - 50+ small, specialized models fine-tuned for key tasks in enterprise process automation (fact-based question-answering, classification, summarization, and extraction). https://lnkd.in/en5iqcHZ

❄️✨ Claude 3.5 Sonnet is now available in Snowflake Cortex AI! ✨❄️ Anthropic’s latest Claude 3.5 Sonnet model is available within the Snowflake security and governance perimeter. No integrations to setup or manage❗️ Claude's advanced language models will further enhance how developers can build agents that can run ad hoc analytics, extract answers from documents and other knowledge bases and execute other multistep workflows. This also mean that you can build applications within the security perimeter of Snowflake with the models next to your governed data.

Orchestra now integrates with Nimble to change the way businesses look to leverage the internet as a data source. Web scraping has never been as easy, powerful or AI enabled. With Orchestra, there is minimal learning curve required to make full-use of Nimble’s AI-web scraping abilities, in stark contrast to trying to scrape the web at scale using a combination of Beautiful Soup, Selenium, and own-managed infrastructure. Check it out here: https://lnkd.in/dSw_KSPy #dataengineering #generativeai #aipipelines #multimodal

AI is increasingly being packaged in containers, like other forms of software. 📦 In our webinar on November 6 at 1 pm ET, "AI + Containers: Better Together", we'll dive into how AI and machine learning developers and data scientists feel about containers. We'll also explore pressing challenges teams are facing in this area, and how they can be remediated, such as: 🎈 Managing bloated containers 💻 Addressing security concerns 🔗 Navigating dependency issues And more! Save your seat today: https://lnkd.in/g_KBksXn