scale

Winsage
June 19, 2026
Microsoft has introduced the Microsoft Execution Containers (MXC) SDK to establish Windows as a reliable operating system for autonomous agents, focusing on containment, identity, and manageability. The MXC framework serves as a policy-driven execution layer for agents on Windows and Windows Subsystem for Linux (WSL), allowing developers to set access permissions using JSON or TypeScript. It employs process and session isolation for agent containment and identity. Future enhancements will include micro-VM support for high-risk tasks and integration with Windows 365 for cloud PC workloads. IT teams can manage MXC policies through Entra ID and Intune, while Defender and Purview provide protection and observability. The MXC framework is built on Microsoft's security initiatives, including Secure Boot and passwordless sign-in, allowing agents to inherit a secure foundation. However, early commentary expresses caution regarding MXC's perception as a comprehensive security solution, noting issues with overly permissive policies and the lack of outbound network filtering. Other platforms, such as Linux, are also enhancing security for agents with kernel-level isolation and secure environments like NVIDIA's OpenShell runtime. Various projects are focusing on agent sandboxes within Kubernetes, employing technologies like gVisor and Kata Containers for isolation. Overall, no singular dominant platform security model for AI agents has emerged, with Windows' MXC still considered nascent compared to existing solutions in Linux and Kubernetes ecosystems.
Winsage
June 18, 2026
Former Microsoft engineer Dave Plummer reflected on software performance challenges during the '90s, specifically the limitations of machines with 12 MB of RAM. He discussed the Basic Block Tool (BBT), which optimized software binaries by reorganizing their structures to enhance execution speed. A typical binary contained around 10 MB of code, but only about 300 KB was necessary for startup. If this essential code was scattered, it led to performance degradation due to excessive page access. BBT defragmented binaries, grouping related code to streamline access and improve user experience, particularly for large products like Windows and Office. Modern tools like BOLT and HP's Dynamo continue to optimize binary layouts and runtime code, but Plummer cautioned about the risks of manipulating binaries. He noted that while computational power has increased, modern software faces similar performance challenges, with larger binaries and complex dependency graphs. He emphasized the importance of locality in software design, advising to keep hot data and code together to enhance performance.
Tech Optimizer
June 18, 2026
Microsoft's Build event highlighted its new AI agent, Scout, while SQL Server received limited attention, raising concerns about its future following Rohan Kumar's departure. Arun Ulag now oversees SQL Server, but analysts note a shift in priorities with SQL Server seemingly less emphasized. The 2022 SQL Server release was viewed as more of a marketing effort than a response to customer needs. Despite the introduction of vector search in SQL Server 2025, competitors had already offered similar features. Microsoft is shifting towards open-source solutions and PostgreSQL, although it reassured users of its commitment to SQL Server. SQL Server, launched in 1989, remains popular, ranking behind Oracle and MySQL. The on-premises database market is lucrative, generating significant revenue, and SQL Server holds a substantial share. Microsoft is unlikely to abandon this profitable segment, aiming to transition users to Azure SQL and SQL database within Fabric. However, migration compatibility issues may arise. Microsoft is also investing in PostgreSQL offerings to compete in the cloud database market, which is evolving rapidly. AWS currently leads in cloud DBMS revenue, posing a challenge for Microsoft. Despite uncertainties, support for SQL Server 2025 is guaranteed until 2036.
Tech Optimizer
June 18, 2026
Lakebase Search is a hybrid vector and full-text retrieval system integrated into Lakebase, now in beta on AWS and Azure. It utilizes two Postgres extensions: lakebase_vector and lakebase_text, allowing agents to operate on a single data backend. Agents manage four times more databases than human users and require real-time access to indexed data. The system features a tiered architecture that stores cold data in cost-effective object storage while keeping active data in local NVMe, significantly reducing costs. The lakebase_vector extension offers 32x compression for vectors, allowing a billion vectors to fit into under 10GB of RAM. The lakebase_text extension provides BM25 relevance ranking without high RAM usage. Benchmarking shows that Lakebase Search can efficiently handle large-scale workloads, achieving high recall and low latency with reduced resource requirements compared to traditional architectures. The system allows for continuous search experimentation and dedicated retrieval engines for each agent, enhancing operational efficiency and scalability.
Tech Optimizer
June 18, 2026
Organizations are increasingly adopting EnterpriseDB's EDB Postgres AI platform due to a rising demand for enhanced control over data in AI systems, particularly in sectors like banking, insurance, retail, and trading. Research from MIT Technology Review Insights indicates that prioritizing AI and data sovereignty significantly predicts success in AI initiatives, with such organizations achieving five times the return on investment. In South Korea, the Industrial Bank of Korea migrated 15 core systems to EDB Postgres AI, citing significant reductions in licensing costs compared to Oracle and improved scalability for future AI services. Shinhan EZ Insurance transitioned its core system to the public cloud using EDB, overcoming challenges related to legacy database licensing and emphasizing the importance of operational stability. Beyond finance, companies like MNTN, Euronext FX, and Kyobo Book Centre have adopted the platform to reduce vendor reliance, manage data workloads, and enhance compliance control. MNTN uses EDB for large-scale analytical processing, Euronext FX has implemented it across four data centers, and Kyobo Book Centre migrated from a costly data warehouse to the EDB solution. A common trend among these deployments is the use of a single Postgres-based platform for transactional processing, analytics, and AI tasks, reflecting an industry initiative to simplify operations and reduce costs. Hensley noted the critical convergence of AI systems with operational data, as AI agents operate against live data in high-volume workflows, highlighting the drawbacks of using separate platforms for transactions and analytics. EnterpriseDB has also received industry accolades for its data management and contributions to the open-source community, reinforcing its market position.
AppWizard
June 18, 2026
Prologue: Go Wayback! has exited early access on Steam, but is considered incomplete with no further development plans. Launched in early access in 2025 by PlayerUnknown Productions, the game faced significant challenges, including staff layoffs, leading to a halt in development. The studio, founded by Brendan Greene, aimed to use the game as a testing ground for its Melba game engine, designed for creating large procedural worlds. A free tech demo, Preface: Undiscovered World, is available in early access. The final version of Prologue includes new features such as enhanced paths, mobile weather monitors, and improvements in cooking, lighting, clouds, and fog.
AppWizard
June 17, 2026
Onimusha: Way of the Sword features a demo that showcases a later segment of the game set in Kyoto's open areas, highlighting its art direction, character movements, and swordplay. The melee combat differs from FromSoftware titles, offering immersive animations and impactful feedback. The demo was perceived as easy, particularly in "Action" mode, with basic enemies posing little threat, while the challenge increased during the boss fight. The game's director, Nihei, noted that both the demo and the event version are early in the game's progression, explaining the ease of basic enemies and indicating that players will face tougher foes as they advance. The demo included button prompts over enemies, which will not be present in the final product. Capcom intends to gradually introduce Musashi's abilities and increase enemy difficulty in the final release. The game will launch with only two difficulty settings: Story and Action, on September 24.
Tech Optimizer
June 17, 2026
Databricks has introduced Lakebase Search, a feature that integrates advanced search capabilities into its Lakebase Postgres database, currently in beta on AWS and Azure. This feature aims to enhance AI agent development by embedding native retrieval functions within the data backend. It addresses the challenge of "Vector Bloat Cost" by utilizing tiered storage for optimized data access and retrieval efficiency. Lakebase Search includes two new Postgres extensions, lakebase_vector and lakebase_text, which enable hybrid search capabilities that combine vector and full-text search functionalities. This integration streamlines the AI agent loop, improving agent-first ergonomics and allowing developers to create more efficient AI systems.
AppWizard
June 17, 2026
The second installment of Steam Next Fest for 2026 is running until June 22 and features a total of 8,682 demos, nearly double the entire library of PlayStation 2 games. Approximately 1,700 of these demos incorporate AI technology, leading to concerns about the quality of some titles, often labeled as "shovelware." Various digital storefronts are struggling to maintain quality amidst the influx of AI-generated content, with Sony recently removing a publisher known for low-quality releases. The reaction to AI in game development has been mixed, with backlash against certain titles. Additionally, early Geekbench results for Valve's upcoming Steam Machine suggest a potential launch next week.
Search