AI engineering article archive

Explore guides and articles on AI development, retrieval systems, and production engineering. Page 6 of 11.

• Architecting Mvp AI

  AI can’t replace architects but helps teams make better decisions faster when building MVPs under pressure, especially through experimentation and context-driven support.

• Spring Ai 1 0

  Jakarta EE 11 introduces support for Java 21, records, virtual threads, and Jakarta Data, laying the groundwork for Jakarta EE 12 and its emphasis on unified data access across persistence APIs.

• Green Microservices

  Microservices often consume more energy than monoliths due to distributed overhead. Architects can make design decisions that improve sustainability.

• Jakarta Ee 11 Overview

  Jakarta EE 11 introduces support for Java 21, Java records, virtual threads, and Jakarta Data, laying the groundwork for Jakarta EE 12 with a focus on unified data access across persistence APIs.

• Guide Curating Technical Conference Track

  One first-time track host shares the process, constraints, and takeaways from building a track from scratch at QCon London 2025.

• Reproducible Ml Iceberg

  Apache Iceberg and SparkSQL bring database-like reliability to your data lake. Time travel, schema evolution, and ACID transactions help support reproducible machine learning experiments.

• Backend Finops Cost Efficiency

  Backend FinOps integrates financial discipline into microservices design and operations. Early adoption, language choice, robust tagging, and automation significantly cut cloud costs.

• Ai Agent Cli

  Well-designed CLIs are crucial in the agentic AI era—serving both human users and autonomous agents with precision and reliability.

• Aws Sandbox As A Service

  Transform AWS cloud experimentation with a secure, automated sandbox framework that minimizes costs, enforces governance, and empowers innovation through efficient environment management.

• One Network Service Policy Architecture

  At QCon SF 2024, Anna Berenberg detailed One Network, which unifies policy and networking across clouds, runtimes, and environments for consistent, scalable management.

• Infusing Ai Java

  Equip yourself with the basic AI knowledge and skills you need to start building intelligent and responsive Enterprise Java applications with Java frameworks like LangChain4j with Quarkus.

• Zero Downtime Cloud Upgrades

  Critical cloud upgrades at scale require systematic preparation to avoid common mistakes. The article covers solutions for legacy systems, performance validation, and rollback plans.

• Ai Infrastructure Aggregating Agentic Traffic

  In this article, author Eyal Solomon discusses AI Gateways, the outbound proxy servers that intercept and manage AI-agent-initiated traffic in real time to provide centralized policy enforcement.

• Ransomware Resilient Storage Cyber Defense

  How ransomware defense has evolved from traditional antivirus to a focus on the storage layer, using immutable backups, AI-powered detection, and isolated vaults to ensure data integrity.

• Virtual Think Tank

  Using LLMs as a virtual think tank enables architects to simulate multi-perspective debates, evaluate trade-offs, and refine decisions—without convening costly expert panels.

• Mcp Connector For Building Smarter Modular Ai Agents

  In this article, the authors discuss Model Context Protocol (MCP), an open standard designed to connect AI agents with tools and data they need, how MCP empowers agent development, and its adoption.

• Kotlin Multiplatform Evaluation

  KMP is emerging as an alternative for cross-platform development, offering a path to share code without sacrificing the performance and feel of a native application, with its own trade-offs.

• Causal Reasoning Observability

  LLMs excel at summarizing observability data but struggle with root cause analysis. This article argues that causal reasoning with Bayesian inference provides more reliable incident diagnosis.

• Architecting Rag Pipeline

  Hybrid search, smart chunking, and domain-aware indexing are key to building effective RAG pipelines. Context window limits and prompt quality critically affect LLM response accuracy.

• Software Engineers Excel AI

  This panel explores how artificial intelligence is reshaping software development, and how software developers and engineering leaders need to become adaptable and resilient in the age of AI.

• Tls Certificate Transparency

  Certificate Transparency logs all TLS certificate issuances publicly, enabling detection of malicious or mistaken certificates that traditional certificate authorities couldn't prevent.

• Federated Credentials Management W3c Proposal

  FedCM is a proposed standard API for frictionless, privacy-preserving federated login on the web that simplifies login for users and developers. FedCM is already supported in Chromium browsers.

• Timeseries Db Rust

  Paul Dix, CTO of InfluxData, describes the Rust rewrite of the engine around Apache Flight, Data Fusion, Apache Arrow and Parquet (the FDAP stack).

• AI & Data Engineering Trends 2025: Agents, Protocols & Delivery

  Review InfoQ’s 2025 AI trends: physical AI, software co-creation, AI-assisted DevOps and agent interoperability through MCP and A2A.

• Llms Evolution Ai Infrastructure

  Disaggregated architectures revolutionize large language model infrastructure by optimizing the prefill and decode phases, delivering throughput improvements and better resource utilization.

• Pipeline Language Migrations

  A pipeline-based approach enables automated, idiomatic code migrations by breaking the process into verifiable stages, avoiding risky rewrites and preserving business logic.

• Onnx Ai Inference With Java

  Bring transformer-based AI into Java with ONNX—no Python required. Learn how to run models natively in the JVM with full support for tokenizers, GPU acceleration, and secure, modular deployment.

• Unintended Consequences Automation Software

  This article lays out some common assumptions and misconceptions about automation, with ideas around how people can better design automated tools to help people better handle software incidents.

• Go Channels Happens Before Concurrency

  This article explores the happens-before semantics of Go channels and how it relates to memory visibility, synchronization, and correctness. We examine pitfalls and explore architectural implications.

• Architecture Quote Your Boss

  Software architecture mirrors organizational dynamics. Misalignment, not bad code, causes architectural failure. Architects must enable organizations to design better systems.

• PDCA AI Code Generation

  AI code generation tools often create quality issues, integration problems, and delivery delays. A structured Plan-Do-Check-Act cycle can maintain code quality while leveraging AI capabilities.

• Three Questions Better Architecture

  MVP success hinges on three questions: Is the idea worth building? Can it scale and perform? Is it maintainable? Use experiments to guide Minimum Viable Architecture decisions.

• Cloud Devops Trends 2025

  InfoQ editorial staff and friends of InfoQ are discussing the current trends in the domain of Cloud and DevOps as part of the process of creating our annual trends report.

• Rag With Spring Mongo Open Ai

  Discover how Spring Boot, Spring AI, MongoDB Atlas and OpenAI can create scalable, efficient solutions by combining business data and generative AI to produce intelligent, contextualised responses.

• Generative Testing

  Generative testing uncovers hidden software bugs by exploring the input space and verifying system invariants. This surpasses example-based tests that rely on known scenarios and can miss edge cases.

• Empowering Decentralizing Architectural Decision Making

  Through stories from our own journey, we’ll share how decentralizing decisions improved alignment across teams, empowered faster decision-making, and fostered a culture of ownership.

• Training Data Preprocessing For Text To Video Models

  In this article, author Aleksandr Rezanov discusses the data preparation for generative text-to-image models to accelerate work on video generation services to be used in TV series and films.

• Building Resilient Platforms Mission Critical Infrastructure

  Matthew Liste, Executive Vice President and Global Head of Infrastructure, presents insights from over 20 years in building mission-critical infrastructure platforms primarily for financial services.

• Scaling Reverse Proxies

  Operating massive reverse proxy fleets reveals hard truths: optimizations fail at different scales, mundane oversights cause outages, and abstractions hide fragility.

• Reducing False Positives Retrieval Augmented Generation

  In this article, author Elakkiya Daivam discusses why Retrieval Augmented Generation (RAG) and semantic caching techniques are powerful levers for reducing false positives in AI powered applications.