InfoQ Homepage Architecture & Design Content on InfoQ
-
JDK 26 and JDK 27: What We Know So Far
JDK 26, the first non-LTS release since JDK 25, has reached its second release candidate with a final set of 10 new features, in the form of JEPs, that can be separated into five categories: Core Java Library, HotSpot, Java Language Specification, Security Library and Client Library. We examine JDK 26 and predict what features have, or could be, targeted for JDK 27.
-
Reducing Onboarding from 48 Hours to 4: inside Amazon Key’s Event-Driven Platform
Amazon Key modernized its event platform by adopting a centralized, event-driven architecture built on Amazon EventBridge. The redesign processes millions of daily events with millisecond latency, improves schema governance, automates cross-account routing, and reduces service onboarding time from 48 hours to four, while maintaining 99.99 percent reliability.
-
How Dropbox Built a Scalable Context Engine for Enterprise Knowledge Search
Dropbox engineers have detailed how the company built the context engine behind Dropbox Dash, revealing a shift toward index-based retrieval, knowledge graph-derived context, and continuous evaluation to support enterprise AI at scale.
-
Uber and OpenAI Retool Rate Limiting Systems
Uber and OpenAI are replacing static rate limits with adaptive, infrastructure-level platforms. Uber’s Global Rate Limiter utilizes probabilistic shedding to manage 80M RPS, while OpenAI’s Access Engine implements a credit waterfall to prevent user interruptions. Both architectures utilize distributed enforcement and soft controls to maintain system stability and service continuity at scale.
-
OpenAI Publishes Codex App Server Architecture for Unifying AI Agent Surfaces
OpenAI has recently published a detailed architecture description of the Codex App Server, a bidirectional protocol that decouples the Codex coding agent's core logic from its various client surfaces. The App Server now powers every Codex experience, including the CLI, the VS Code extension, and the web app, through a single, stable API.
-
GlassFish 8.0 Delivers Compatibility with Jakarta EE 11, Enhanced Security and Improved Data Access
After 15 milestone releases, the Eclipse Foundation has released version 8.0 of GlassFish featuring support for virtual threads; enhanced application security; and improved data access. GlassFish 8.0 is a compatible implementation of Jakarta EE 11. Ondro Mihályi, Java Champion and co-founder of OmniFish, spoke to InfoQ about GlassFish 8.0.
-
Agoda’s API Agent Converts Any API to MCP with Zero Code and Deployments
Agoda engineers developed API Agent, enabling a single MCP server to access any internal REST or GraphQL API with zero code and zero deployments. The system reduces overhead from multiple APIs, supports AI-assisted queries, and uses in-memory SQL post-processing for safe, scalable data handling across internal services.
-
Google Explores Scaling Principles for Multi-Agent Coordination
Google Research tried to answer the question of how to design agent systems for optimal performance by running a controlled evaluation of 180 agent configurations. From this, the team derived what they call the "first quantitative scaling principles for AI agent systems", showing that multi-agent coordination does not reliably improve results and can even reduce performance.
-
Java News Roundup: JDK 26-RC1, Spring Framework, Open Liberty, EclipseStore, Gradle
This week's Java roundup for February 9th, 2026, features news highlighting: the first release candidate of JDK 26 and Gradle 9.4; milestone releases of Micrometer Metrics and Micrometer Tracing; beta releases of Open Liberty 26.0.0.2 and EclipseStore 4.0; and maintenance releases of Spring Framework, Micronaut, and Quarkus.
-
Firestore Adds Pipeline Operations with over 100 New Query Features
Google has overhauled Firestore’s query engine, introducing "Pipeline operations" that enable complex server-side aggregations and array unnesting. The update shifts Firestore Enterprise toward an optional indexing model, allowing architects to prioritize write speed and lower costs. While it brings parity with MongoDB-style aggregations, the preview currently lacks real-time and emulator support.
-
OpenAI Scales Single Primary PostgreSQL Instance to Millions of Queries per Second for ChatGPT
OpenAI described how it scaled PostgreSQL to support ChatGPT and its API platform, handling millions of queries per second for hundreds of millions of users. By running a single-primary PostgreSQL deployment on Azure with nearly 50 read replicas, optimizing query patterns, and offloading write-heavy workloads to sharded systems, OpenAI maintained low-latency reads while managing write pressure.
-
WhatsApp Deploys Rust-Based Media Parser to Block Malware on 3 Billion Devices
WhatsApp has rewritten its media handling library in Rust, replacing 160,000 lines of C++ with 90,000 lines of memory-safe code for 3 billion devices. The rollout, part of a system called Kaleidoscope, uses differential fuzzing to ensure bug-for-bug compatibility. The move mirrors a decade-long industry shift toward memory safety, tracing back to Mozilla's first Rust MP4 parser deployment in 2016.
-
CloudFront Adds Origin mTLS Authentication for End-to-End Zero Trust
Amazon CloudFront now supports mutual TLS authentication for origin servers, completing end-to-end zero-trust authentication from viewers to backends. The feature replaces IP allowlists and shared secrets with cryptographic verification, proving particularly valuable for multi-cloud deployments, where origins can verify that traffic originated from CloudFront without VPN tunnels.
-
QCon Previews 20th Anniversary Conferences: Production AI, Resilience, and Staff+ Engineering
Celebrating its 20th anniversary, QCon’s 2026 conferences in London and San Francisco will focus on the engineering realities of agentic AI, resilient architectures, and platform ROI. The programs continue the series' two-decade tradition of practitioner-led content, curated by senior engineers from companies like Zoox, UBS, and LinkedIn.
-
Uber Moves In-House Search Indexing to Pull-Based Ingestion in OpenSearch
Uber transitions its in-house search indexing to OpenSearch with a pull-based ingestion framework, improving reliability, backpressure handling, and multi-region consistency for large-scale streaming data while simplifying recovery and supporting global, real-time search experiences.