InfoQ Homepage News
-
Dreamer 4: Learning to Achieve Goals from Offline Data Through Imagination Training
Researchers from DeepMind have described a new approach for teaching intelligent agents to solve complex, long-term tasks by training them exclusively on video footage rather than through direct interaction with the environment. Their new agent, called Dreamer 4, demonstrated the ability to mine diamonds playing Minecraft after being trained on videos, without ever actually playing the game.
-
Uber Achieves 150M Reads per Second with CacheFront Improvements
Uber has updated its CacheFront architecture to handle over 150 million reads per second. The new design improves consistency and reduces stale reads by integrating Flux for MySQL binlog tailing, enhancing the storage engine, and introducing Cache Inspector for monitoring and optimization.
-
Java News Roundup: JDK 26 Release Schedule, GlassFish, Spring AI, Helidon, Mandrel 25, Seed4j
This week's Java roundup for September 29th, 2025, features news highlighting: JEP 527, Post-Quantum Hybrid Key Exchange for TLS 1.3; the proposed JDK 26 release schedule; the release of Mandrel 25 and Seed4j; milestone releases of GlassFish 8.0 and 7.1; the beta release of WildFly 38; and point releases of Spring AI, Helidon, Micronaut, LangChain4j, Gatherers4j and Keycloak.
-
New DNS Armor Service Helps Google Cloud Workloads Preemptively Block Cyber Threats
Google Cloud's DNS Armor, in partnership with Infoblox, offers a vital layer of security against DNS-based threats for Google Cloud workloads. Utilizing advanced threat detection and machine learning, it identifies and mitigates risks like malware and data exfiltration, ensuring robust protection without impacting performance. Deployable as a managed service providing seamless control for users.
-
Vercel Introduces Drains for Unified Data Export
Vercel has released Vercel Drains, a system for exporting observability data from its platform into external services. The feature unifies logs, distributed traces, web analytics events, and performance metrics into a single streaming mechanism.
-
AWS Introduces M4 and M4 Pro Mac Instances for Faster Apple App Development
AWS has recently launched two new Mac instances (M4 and M4 Pro) built on Apple's latest M4 silicon. The new EC2 instances provide faster CPU performance, enhanced graphics, and increased memory for building iOS and macOS applications.
-
Agoda Leverages ChatGPT in the CI/CD Process for SQL Stored Procedure Optimization
Agoda started utilizing ChatGPT to optimize SQL stored procedures (SP) as part of their CI/CD process. After introducing the automated LLM-assisted step, the company observed shortened stored procedure optimization times, which lightened the load on DB developers. Agora works on making ChatGPT more accessible for SP optimization outside of the CI/CD pipeline.
-
Anthropic Reveals Three Infrastructure Bugs behind Claude Performance Issues
Anthropic recently published a postmortem revealing that three distinct infrastructure bugs intermittently degraded the output quality of its Claude models in recent weeks. While the company states it has now resolved those issues and is modifying its internal processes to prevent similar disruptions, the community highlights the challenges of running the service across three hardware platforms.
-
How Software Engineers Can Grow into Staff Plus Roles
Software engineers can boost their impact by helping other teams, focusing on business-driven work, and building strong relationships, David Grizzanti mentioned at InfoQ Dev Summit Boston. Growth can come from mentoring, setting cultural norms, thinking strategically, and designing a career path based on what motivates you.
-
Microsoft Announces General Availability of AKS Automatic
Microsoft has released Azure Kubernetes Service (AKS) Automatic to general availability, introducing a fully-managed Kubernetes offering designed to eliminate operational overhead while maintaining the full power and flexibility of the platform.
-
Microsoft Announces Open-Source Agent Framework to Simplify AI Agent Development
Microsoft has announced the preview release of Microsoft Agent Framework, an open-source software development kit designed to simplify the creation and deployment of artificial intelligence agents for developers across all skill levels, as reported in official blog posts from the company's development teams.
-
Microsoft Tests Microfluidic Cooling for Next-Generation AI Chips
Microsoft has announced progress on a new chip cooling approach that could help address one of the biggest bottlenecks in scaling AI infrastructure: heat. The company’s researchers have successfully demonstrated in-chip microfluidic cooling, a system that channels liquid coolant directly into etched grooves on the back of silicon chips.
-
Cursor 1.7 Adds Hooks for Agent Lifecycle Control
Cursor has introduced a Hooks system in version 1.7 that allows developers to intercept and modify agent behavior at defined lifecycle events. Hooks can be used to block shell commands, run formatters after edits, or observe agent actions in real time.
-
Perplexity Launches Search API to Power Next-Gen AI Applications
Perplexity has introduced the Search API, opening up access to the same infrastructure that underpins its public answer engine. With coverage of hundreds of billions of webpages and infrastructure tuned for AI-heavy workloads, the new API is aimed at developers who want real-time, reliable search results for building their own agents, applications, and retrieval-augmented pipelines.
-
Pulumi Launches Neo: an Agentic AI Platform Engineer for Multi-Cloud Infrastructure
Infrastructure automation company Pulumi has introduced what's claimed to be the first artificial intelligence-based platform engineering agent for the industry, named Neo. The tool works to resolve some of the infrastructure bottlenecks that develop as a side effect of AI tools speeding up software development.