Adding more AI-generated code won’t help if your team can’t review or deploy…
Adding more AI-generated code won’t help if your team can’t review or deploy it any faster. Lizzie Matusov (Co-Founder & CEO at Quotient) explains why AI token usage and code volume are poor proxies f…
Generating code with AI is easy. Trusting it in production is the hard part.
Generating code with AI is easy. Trusting it in production is the hard part. In this #InfoQ talk, #Roblox Engineering Manager Andrew Swerdlow explains how his team moved from Copilot autocomplete to a…
Microsoft is shifting AI governance from static policies to runtime enforcement.
Microsoft is shifting AI governance from static policies to runtime enforcement. Its framework spans 9 governance domains and 4 functions - policy, control, visibility & proof - to ensure governance r…
"LLMs aren’t going to understand what an embedding means. These are just…
"LLMs aren’t going to understand what an embedding means. These are just numbers". Sudeep Das (Head of ML/AI at DoorDash) breaks down why agentic recommendation systems are shifting from numerical exp…
#NextJS 16.3 is here with significant improvements:
#NextJS 16.3 is here with significant improvements: ➡️ Lower memory usage ➡️ Faster builds and type checking ➡️ Instant Navigations for faster, client-like responses with server rendering Details here 👉 …
In this #InfoQ article, Srikanth Mamidala explores data lake architecture for…
In this #InfoQ article, Srikanth Mamidala explores data lake architecture for analytics, reporting, and ML, and how to manage consumer lag when using Kafka and Apache Hudi. 🔗 Read now: https://www.inf…
#Cloudflare is turning engineering standards from passive documentation into…
#Cloudflare is turning engineering standards from passive documentation into actively enforced controls across the software development lifecycle. Since early 2026, its AI reviewer has flagged nearly …
Are AI code reviewers outperforming humans?
Are AI code reviewers outperforming humans? At #QConAI Boston 2026, Roblox’s Andrew Swerdlow shared a striking stat: 🔻 Human code review acceptance: ~55% 🟩 Autonomous AI agent acceptance: 68–70% The k…
#DeepSeek just released DeepSeek Harness (dsh) under the MIT license! Built on…
#DeepSeek just released DeepSeek Harness (dsh) under the MIT license! Built on Cordis, it unbundles agent infrastructure into a modular micro-kernel where models, tools, and execution loops operate as…
Cloudflare automated issue triage for the Astro open-source framework with…
Cloudflare automated issue triage for the Astro open-source framework with isolated AI agents running in GitHub Actions. The agents reproduce bugs, diagnose root causes, verify behavior, propose fixes…
Shift-left & DevOps have accelerated software delivery - but also increased…
Shift-left & DevOps have accelerated software delivery - but also increased cognitive load and duplicated effort across testing, security, and maintenance. John Keates explores how to rightsize develo…
The biggest surprise with agentic coding isn’t faster shipping... it’s…
The biggest surprise with agentic coding isn’t faster shipping... it’s Parallel Shipping! Engineers at #OpenAI are now running 7 to 10 concurrent workstreams at once using AI agents. Software developm…
What’s new in .NET 11 Preview 7?
What’s new in .NET 11 Preview 7? Microsoft’s latest preview brings updates across C#, ASP.NET Core, .NET MAUI, Entity Framework Core, Windows Forms, and F#. Explore the changes ⇨ https://bit.ly/3SmWuU…
Why fetch when you can sync?
Why fetch when you can sync? In this presentation on #InfoQ, James Arthur explores local-first sync engine architecture and how to adopt these patterns using TanStack DB: a reactive client store for b…
💡 Dive into our in-depth analysis of #JDK27 and discover what features could…
💡 Dive into our in-depth analysis of #JDK27 and discover what features could be targeted next for #JDK28. Read more on #InfoQ 👉 https://bit.ly/3UCZ0qx #Java #ProjectAmber #ProjectLoom #ProjectPanama #…
"Is this a proof of concept, or will it get me out of bed at 2 a.m.?"
"Is this a proof of concept, or will it get me out of bed at 2 a.m.?" Before letting an #AICoding agent run unsupervised, you need to know your feedback loops. The higher the risk of failure, the tigh…
#SoftwareArchitecture isn’t something you “get right” once. It’s a living…
#SoftwareArchitecture isn’t something you “get right” once. It’s a living Sociotechnical Craft shaped by team structure, friction, flow - and code. Explore the #InfoQ eMag on #EvolutionaryArchitecture…
#Netflix open-sourced an agentic workflow for Observational Causal Inference…
#Netflix open-sourced an agentic workflow for Observational Causal Inference (OCI) that automates parts of causal analysis. Given observational data & a human-defined analysis plan, the agent uses an …
Distributed system failures often mirror structural engineering collapses: a…
Distributed system failures often mirror structural engineering collapses: a single local weakness can trigger a catastrophic cascading failure. Drawing on real-world examples, Sam Newman examines the…
Early-stage engineering culture is shaped far more by founder tendencies than…
Early-stage engineering culture is shaped far more by founder tendencies than formal policy. David Gudeman (Co-Founder & CTO at Velocity AI) talks about the hidden operational costs of these founder t…
InfoQ@techhub.social
·
