Deterministic rules protect hard metrics. But how do you govern architectural…
Deterministic rules protect hard metrics. But how do you govern architectural intent? Agentic fitness functions use AI agents & versioned rubrics to evaluate judgment-heavy concerns like boundary fide…
Is Platform Engineering just DevOps with a new badge?
Is Platform Engineering just DevOps with a new badge? Shweta Vohra shares a surprising reaction from a recent talk: 70% of the room said Platform Engineering is simply a rename of DevOps. 🔹 2010: "We …
#PromptEngineering taught us to write better instructions. #ContextEngineering…
#PromptEngineering taught us to write better instructions. #ContextEngineering asks a bigger question: what should your AI agent see - and what should it never see? In this #InfoQ talk, Patrick Debois…
#Netflix migrated most of its batch workloads to Kueue - an #opensource…
#Netflix migrated most of its batch workloads to Kueue - an #opensource cloud-native batch job execution system, replacing its homegrown solution. By mapping existing capabilities to Kueue and adoptin…
Once you run out of AI tokens - is the workday over?
Once you run out of AI tokens - is the workday over? Matt Saunders highlights the funny - and expensive - reality of modern AI adoption: “Once we’ve used our token budget, should we just go home? Is o…
#JetBrains centralised AI usage after development-related spending increased…
#JetBrains centralised AI usage after development-related spending increased roughly tenfold in 6 months. Instead of limiting engineers to approved tools, it built a shared access and accounting layer…
It’s been a busy week in the #Java ecosystem!
It’s been a busy week in the #Java ecosystem! Highlights include: ➤ JEP 535, Shenandoah GC: Generational Mode by Default, targeted for JDK 28; point releases of A2A Java SDK, Apache Camel and Gradle; …
JEP 401, integrated into #JDK28, introduces value objects to #Java.
JEP 401, integrated into #JDK28, introduces value objects to #Java. Value objects use final fields, different equality semantics, and stricter construction and synchronization rules, aiming to improve…
#Pinterest has revealed the Resource Provisioner Pipeline (RPP) - its own…
#Pinterest has revealed the Resource Provisioner Pipeline (RPP) - its own Terraform execution engine. RPP enforces least-privilege access and dual-control reviews, adding strict guardrails to GitHub A…
What’s shaping #CloudComputing and #DevOps in 2026?
What’s shaping #CloudComputing and #DevOps in 2026? Our annual #TrendsReport breaks down the key topics architects & technical leaders should prioritize to stay ahead of the curve. 📖 Read the full rep…
How Netflix redesigned its Service Topology pipeline to scale real-time…
How Netflix redesigned its Service Topology pipeline to scale real-time service mapping: 🔹 Separates resolution from enrichment & persistence 🔹 Propagates backpressure to Kafka (zero record drops) 🔹 S…
#Angular v22 is here with API stabilizations, ergonomic templates, and new…
#Angular v22 is here with API stabilizations, ergonomic templates, and new tooling for AI coding agents. The release also moves Signal Forms, Angular Aria, and asynchronous reactivity APIs to producti…
AI inference is expensive - but it doesn’t have to be.
AI inference is expensive - but it doesn’t have to be. In this #InfoQ talk, Meryem Arik breaks down how to systematically reduce cost per token across AI workloads, with real-world examples from data …
Find out how #Stripe automated database incident recovery by modeling its…
Find out how #Stripe automated database incident recovery by modeling its global infrastructure as a graph. Using graph search algorithms and state machines, the team automatically computes and execut…
As AI commoditizes code output, system comprehension can silently decay -…
As AI commoditizes code output, system comprehension can silently decay - creating cognitive debt that threatens safe architectural evolution. This #InfoQ article explores why human understanding must…
How does #ChatGPT scale to 900M weekly active users while keeping up with…
How does #ChatGPT scale to 900M weekly active users while keeping up with rapid AI development? Hint: It’s not just a GPU problem. In this #InfoQ talk, Martin Spier explains how performance engineerin…
Vercel Labs introduced Zero - an experimental systems programming language…
Vercel Labs introduced Zero - an experimental systems programming language built for AI agents rather than human developers. Zero features a dedicated toolchain contract, structured error messages, an…
Spotify developers spend less than 1 hour/day writing new features - the rest…
Spotify developers spend less than 1 hour/day writing new features - the rest goes to dependency updates, framework migrations, and technical debt. To address this, Spotify built Honk - an AI-powered …
AI is completely reshaping engineering teams, developer roles & engineering…
AI is completely reshaping engineering teams, developer roles & engineering culture! The #InfoQ Culture & Methods Trends Report 2026 explores: • AI adoption maturity & risk • The evolving engineering …
Choosing the right AI model isn't your first architectural decision.
Choosing the right AI model isn't your first architectural decision. The real "first fork"? Deciding how your AI system is structured - choosing between skills, sub-agents, or alternative patterns. Pr…
InfoQ@techhub.social
·
