MongoDB adds vector search to self-managed editions to power generative AI apps
While at one end, traditional database players, such as MongoDB, Google, etc., have added vector capabilities, specialty vector databases are adding additional features to make their products more easily consumable by non-experts. When asked about the delay in releasing vector search to the self-managed offerings, Andersen pointed out that it could have been a business […]
Read More
MongoDB launches a new application modernization platform
Components of the platform The new platform will offer a dedicated team of engineers, spread globally, who will oversee and guide enterprises through the entire transformation process, including deployment. A typical modernization journey for an enterprise at MongoDB, according to Kolhar, will include a consulting process that will include discussions on expected pricing, and after […]
Read More
2-agent architecture: Separating context from execution in AI systems
When I first started experimenting with voice AI agents for real-world tasks like restaurant reservations and customer service calls, I quickly ran into a fundamental problem. My initial monolithic agent was trying to do everything at once: understand complex customer requests, research restaurant availability, handle real-time phone conversations and adapt to unexpected responses from human […]
Read More
Making good choices: How to get the best from Python tools
The way to get the best from Python, aka “the lingua franca of AI,” is by knowing both its traps and its treasures. One such treasure is using uv run as a magic launcher; another is a chatbot library that spares you the pain of writing SQL for data queries. Get these updates and more, […]
Read More
Databricks at a crossroads: Can its AI strategy prevail without Naveen Rao?
“Databricks is in a tricky spot with Naveen Rao stepping back. He was not just a figurehead, but deeply involved in shaping their AI vision, particularly after MosaicML,” said Robert Kramer, principal analyst at Moor Insights & Strategy. “Rao’s absence may slow the pace of new innovation slightly, at least until leadership stabilizes. Internal teams […]
Read More
Kotlin 2.2.20 boosts WebAssembly support
Kotlin 2.2.20 has been released, with the latest version of the concise, multiplatform, “pragmatic” programming language bringing its WebAssembly-compilation component, Kotlin/Wasm, into beta. JetBrains announced Kotlin 2.2.20 release on September 10. The release includes a beta version of the Kotlin/Wasm tool for compiling Kotlin code into WebAssembly (Wasm), with improvements to exception handling in JavaScript […]
Read More
Unlocking LLM superpowers: How PagedAttention helps the memory maze
1. Memory fragmentation Internal fragmentation Systems pre-allocate a large chunk of memory for each request, assuming the maximum possible output length (e.g., 2048 tokens). However, if a request only generates a short output, much of that reserved memory goes unused, leading to significant waste. External fragmentation Because different requests reserve chunks of varying sizes, the […]
Read More
.NET 10 moves to release candidate stage
Microsoft’s planned .NET 10 software development platform has reached the release candidate (RC) stage, with highlights including post-quantum cryptography API completion and ASP.NET Core Identity metrics. Announced September 9 and downloadable from dotnet.microsoft.com, .NET 10 Release Candidate 1 is the first of two planned release candidates. The initial RC follows seven previews, with the first […]
Read More
When AI nukes your database: The dark side of vibe coding
One July morning, a startup founder watched in horror as their production database vanished, nuked not by a hacker, but by a well-meaning AI coding assistant in Replit. A single AI-suggested command, executed without a second glance, wiped out live data in seconds. The mishap has become a cautionary tale about “vibe coding,” the growing […]
Read More
Google intros EmbeddingGemma for on-device AI
With the introduction of its EmbeddingGemma, Google is providing a multilingual text embedding model designed to run directly on mobile phones, laptops, and other edge devices for mobile-first generative AI. Unveiled September 4, EmbeddingGemma features a 308 million parameter design that enables developers to build applications using techniques such as RAG (retrieval-augmented generation) and semantic […]
Read More