Yunfeng Yang

In case you missed it, we have now released the second issue of our Confluent Chronicles – a fun and entertaining take on the fundamentals of stream processing, Apache Flink, and Confluent! We had such great feedback on the first issue that we couldn’t just stop there – you can check out and download the latest edition here: https://lnkd.in/gEFxA7af Would love to hear what you think of it!   #ApacheFlink #ApacheKafka #Confluent

39

1 Comment

Join my Data Architecture SNIA session featuring AlloyDB Omni at 11:45 PM PDT today! Discover why a robust data architecture is crucial for successful AI implementations. This session explore strategies to manage, store, and optimize data for maximum performance, efficiency, and scalability. #SNIA, #CMSSummit,#AlloyDBomni, #RAG #vectordatabases #pliops

15

Check out this new SIGMOD '24 scientific paper. Our customers love Amazon MemoryDB for its low latency performance, rich flexible API, large scale (up to 500 nodes per cluster), security, availability (4 9s SLA), and AWS's operational excellence. We designed MemoryDB for 11 9s of durability while still providing in-memory performance. My team created a nice technical scientific paper on MemoryDB's architecture. If you would like to learn about what's under the hood, check out this SIGMOD '24 paper. Yacine Taleb, Xingbo Wang, Nan Yan, Allen Samuels, Kevin McGehee, Stefan Mueller, and Jean G. - excellent work publishing this paper! https://lnkd.in/gFXPvci8 #aws #memorydb #inmemory #databases #durability #science #sigmod

192

6 Comments

Interested in integrating SageMaker Processing jobs with PrestoDB for efficient batch transforms and frequent model retraining 🚀 ? Releasing a new AWS blog with Amit Arora, Johnny Chivers and Antara Raisa containing an end to end an open source solution that uses Amazon SageMaker Pipelines and PrestoDB. This blog dives into an end-to-end workflow that seamlessly connects SageMaker Processing jobs to your PrestoDB data sources, enabling optimized batch inference and scheduled model retraining. From data fetching and model training to batch transforms and real-time deployment, you can now use this approach for a variety of use cases all through minimal/no-code changes! Do check it out and let me know if you have any questions/feedback. Blog link: https://lnkd.in/evsgxt3P Open-Source GitHub Repository: https://lnkd.in/e8nsWvGN #AWS #Machinelearning #AmazonSageMaker #MLOps

13

1 Comment

ONNX is an open source machine learning (ML) framework that provides interoperability across a wide range of frameworks, operating systems, and hardware platforms. This new blog post shows you how to run ONNX Runtime inference on #AWS Graviton3 and how to configure to use optimized GEMM kernels. With these optimizations, you can see up to 65% better performance for fp32 inference and up to 30% better performance for int8 quantized inference for several natural language processing (NLP) models on Graviton3: https://lnkd.in/giWYk6Hq

116

4 Comments

What makes success stories like Observe.AI so exciting for me is witnessing MongoDB's evolution into an AI-accelerating data platform. For an AI-centric company like Observe.AI, choosing the right data architecture was the star. And they recognised MongoDB's distinct strengths to flex across databases, analytics, search, mobile data management and now vector databases for generative AI workloads. Our powerful indexing and querying capabilities allow Observe.AI's data scientists to quickly access and analyse the data powering their AI models. Read the blog to know how MongoDB is a core enabler of their innovative AI stack. #MongoDB #artificialintelligence #dataanalytics

28

Experience Mixtral 8x22B and Mistral Large API endpoints, powered by NVIDIA NIM microservice, from our NVIDIA API catalog ➡️ https://lnkd.in/gXsNpu2J These performance-optimized #generativeAI models by Mistral AI, excel in complex multilingual reasoning tasks, and more.

4

RAG architectures enhance #LLMs by integrating proprietary data, but maintaining data freshness poses challenges. 🤯 Continuous updates of vector embeddings ensure accuracy in AI models’ responses. ✅ The below solution uses MongoDB Atlas Stream Processing and MongoDB Atlas Vector Search to streamline the updating and searching of embeddings, simplifying the process for developers. 👇 https://lnkd.in/eNGxSp7k

32

1 Comment

Quantization methods allow us to create smaller vector indexes, resulting in cost savings and performance gains as a trade-off for a very small amount of quality loss. Typically, you start with vector embeddings that are lists of 32-bit floats. Quantizing to int8 (1 byte per dimension) has been known to work well in many cases, but I continue to be surprised that binary vectors, where you use a single bit per dimension, perform as well as they do (depends on the embedding model, but they often show surprisingly low quality impact considering they are 32x smaller than the original). In this post Farzad shows how you can use int8 and binary vectors in Azure AI Search, and how you can integrate them with the great work Cohere has done offering float32, int8 and binary vector embeddings directly on their API.

89

1 Comment

#AI is nothing new to #Aerospike. Read more about our company’s recent #funding landmark and future plans to further support our AI and #cloud services in this SiliconANGLE & theCUBE article. #GrowthCompany #NoSQL

1

At Elemental Cognition, we’re transforming #GenAI solutions using #ApacheKafka and Confluent. In this The New Stack article, I shared how our AI platform uses Kafka and Confluent Cloud to stream and vet data in real-time before the EC Reasoning Engine applies formal logic for transparent, accurate solutions. Check out the article for a deep dive into the best practices we’ve learned for building reliable GenAI apps. https://lnkd.in/eU5RupZu

37

4 Comments

We have been closely monitoring software engineering leaders committed to implementing smart and predictive features. However, they often face a skills gap within their teams. #Cloud AI Developer Services (#CAIDS) help bridge this gap by enabling developers who aren't data specialists to utilize #AI models via #APIs and SDKs. These services include automated machine learning, data preparation, feature engineering, and model management, along with language and vision capabilities such as sentiment analysis and image generation. According to the latest Gartner report on cloud AI developer services, Microsoft and AWS are recognized as leaders, exemplifying innovation, and operational efficiency. The study notes that Microsoft, with its rapid response to market demands and its business model, offers affordable solutions through Azure AI, bolstered by strategic partnerships with OpenAI and Hugging Face, thereby strengthening its position by innovating in its product lines. On the other hand, Amazon Web Services (AWS) stands out with Amazon SageMaker and other popular language and vision services, simplifying the AI and ML cycle and enabling developers to create their solutions supported by a global infrastructure.

23

Exactly one month ago, Intuit's Kalyan Kolachala (Director, Product Development), gave a Lightning Talk: Federated Search Over Distributed Observability Data at #ArgoCon | #KubeCon EU. In this talk, he gives insight into the use of technologies like, #OpenTelemetry, #ApacheArrow and #ApachePresto and how it can be used to improve speed, performance and significantly bring down Observability costs (for compute, networking and storage) and illustrates the same using a case study from Intuit. https://lnkd.in/gnv3vESJ #Intuit #IntuitOpenSource #OpenSourceCommunity #Observability #OpenSource #CloudNativeCon

15

With the announcement of S3-native-streams (Freight clusters), here is a commentary on Confluent strategy regarding object storage, streaming and an open data architecture. https://lnkd.in/dTg6DKVy

48

6 Comments

Learn how to analyze audio customer reviews using BigQuery ML's speech-to-text capabilities, and gain valuable insights from unstructured data to improve your business. Explore practical use cases and find a concise guide in this comprehensive blog post: https://lnkd.in/gPux-3Wn

1

Generative AI is changing the data storage landscape as file sizes continue to grow. My latest piece for Fast Company Executive Board on why file sizes are growing and what organizations can do to manage them in a cloud environment. https://lnkd.in/gNDT3-Fh #GenerativeAI #Cloud #ContentManagement

160

3 Comments

Follow this tutorial to learn how to build a retrieval-augmented generation solution using #OCI #GenerativeAI. https://lnkd.in/ek3kSniK

7