ACM Queue

05/27/2026

Eight Myths on Software Engineering and GenAI
Examining the most common misconceptions

Generative AI is reshaping software engineering—but the narrative has gotten ahead of the evidence. Marketing claims, anecdotal wins, and misread studies have given rise to a set of persistent myths that are quietly driving poor decisions about AI adoption, tooling, and how to measure success.

This article examines eight of the most common misconceptions. We already know developers don’t actually spend most of their time writing code, with studies at Microsoft and elsewhere showing it’s closer to 14 percent. That means AI code generation, even when it works well, touches a surprisingly small slice of the actual job. And yet organizations are doubling down on lines-of-code metrics to track AI’s impact, which is a measure that is neither statistically valid nor meaningfully connected to outcomes such as software quality or delivery speed.

The reality is messier and more interesting than the headlines suggest. AI works better for some tasks, some developers, and some contexts than others. Productivity gains don’t flow automatically from handing engineers a license—they require rethinking workflows at the organizational level. Adoption stalls when developers don’t trust the tools, lack time to learn them, or worry about de-skilling. And the “startups move fast with AI” narrative ignores the compliance, legacy systems, and reliability constraints that define enterprise software.

This article isn’t skeptical, but rather provides practitioners, team leads, and engineering leaders a clearer, research-backed picture so the decisions organizations make about AI are grounded in evidence, not just enthusiasm.

Generative AI is reshaping software engineering—but the narrative has gotten ahead of the evidence. Marketing claims, anecdotal wins, and misread studies have given rise to a set of persistent myths that are quietly driving poor decisions about AI adoption, tooling, and how to measure success.

05/19/2026

Climbing the Generative AI Mountain
A “hitchhiker’s guide” for product managers

Most product managers working in software today feel it—that dizzying sense of standing at the base of a mountain, staring up at a new way of working and wondering where to even begin. Generative AI has fast arrived in the software engineering workforce, with little guidance and high expectations. PMs are expected to be more productive almost overnight, without a map for how to get there.

Microsoft has been at the center of this transformation. Drawing on interviews, surveys, and telemetry from 885 PMs on software teams at Microsoft, we studied how practitioners at every stage of the climb are actually navigating this shift—from those still finding their footing at base camp to those who have reached the summit and fundamentally redesigned how they work. From this data, we developed a set of personas that capture the real texture of the ascent: where people get stuck, what moves them forward, and what traps are easy to fall into.

This article is not a feature overview or a list of prompting tips. It is a practitioner’s guide built from real behavior, real struggles, and real wins—with verbatim accounts from PMs doing this work every day. Whether you are still packing your bags or looking to push past the next ridge, this article offers concrete, experience-backed guidance for redesigning your workflows with AI, one step at a time.

Most product managers (PMs) working in software today feel it—that dizzying sense of standing at the base of a mountain, staring up at a new way of working and wondering where to even begin. Generative AI has fast arrived in the software engineering workforce, with little guidance and high expecta...

05/06/2026

The AI-Native Developer
Redefining work, identity, and the future of craft

AI is changing software development in a way that forces a more uncomfortable question: Which parts of the job are still worth doing? Developers are making deliberate choices about what to keep, what to delegate, and what they no longer recognize as their work. Many report that their work feels less meaningful than before, suggesting a deeper shift in the role itself. Drawing on large-scale mixed-methods surveys of developers and in-depth interviews with AI-fluent practitioners, we investigate what it actually means to be a software developer today, how the role evolves as AI fluency deepens, and where this all might lead. We explore what futures become possible as AI augments software creation and what choices might help us design for the futures worth wanting.

AI is changing software development in a way that forces a more uncomfortable question: Which parts of the job are still worth doing? Developers are making deliberate choices about what to keep, what to delegate, and what they no longer recognize as their work. Many report that their work feels less...

04/28/2026

Knowledge Graphs over Two Decades
From web-scale extraction to LLM-augmented intelligence

Xia Luna D**g

This paper traces the evolution of knowledge graphs across three generations: entity-based knowledge graphs (KGs), text-rich KGs, and the emerging convergence of KGs and large language models. The boundary between symbolic and neural knowledge continues to blur, leading to a new era of flexible, context-aware knowledge systems.

Knowledge Graphs (KGs) have powered a wide range of applications, from web search to personal assistants. They have evolved over three generations: entity-based KGs, which support general search and question answering (e.g., Google, Bing); text-rich KGs, which enhance product recommendations, biomedical discovery, and other domain applications (e.g., Amazon, Alibaba); and the emerging convergence of KGs and large language models (LLMs), called dual neural knowledge.

This article analyzes the defining characteristics of each generation, the underlying methods for knowledge integration and extraction, and the technical innovations that have driven their industrial impact. As LLMs advance machine understanding of natural language, the boundary between symbolic and neural knowledge continues to blur, leading to a new era of flexible, context-aware knowledge systems with transformative potential for both research and industry.

This paper traces the evolution of knowledge graphs across three generations: entity-based knowledge graphs (KGs), text-rich KGs, and the emerging convergence of KGs and large language models. The boundary between symbolic and neural knowledge continues to blur, leading to a new era of flexible, con...

04/21/2026

Operations and Life

The Important Decision Document:

Every project (and family) needs one.

Long-running projects are a journey. An IDD (important decisions doc) captures knowledge and decisions gathered along the way. It records the what and why of decisions, as well as the rationale for rejecting alternatives. It helps new team members get up to speed, prevents wasting time on relitigating old decisions, improves morale, and increases accountability.

Long-running projects are a journey. An IDD (important decisions doc) captures knowledge and decisions gathered along the way. It records the what and why of decisions, as well as the rationale for rejecting alternatives. It helps new team members get up to speed, prevents wasting time on relitigati...

04/14/2026

Kode Vicious: KV the Apostate

Faith-based computing versus the unnatural science

Whether we ask an LLM or a recent graduate to type the code is less important than knowing what the code does, how it was built, and when to look under the hood

07/31/2018

Knowledge Base Construction in the Machine-learning Era

Three critical design points: Joint-learning, weak supervision, and new representations

This installment of Research for Practice features a curated selection from Alex Ratner and Chris Ré, who provide an overview of recent developments in Knowledge Base Construction (KBC). While knowledge bases have a long history dating to the expert systems of the 1970s, recent advances in machine learning have led to a knowledge base renaissance, with knowledge bases now powering major product functionality including Google Assistant, Amazon Alexa, Apple Siri, and Wolfram Alpha. Ratner and Ré's selections highlight key considerations in the modern KBC process, from interfaces that extract knowledge from domain experts to algorithms and representations that transfer knowledge across tasks.

This installment of Research for Practice features a curated selection from Alex Ratner and Chris Ré, who provide an overview of recent developments in Knowledge Base Construction (KBC). While knowledge bases have a long history dating to the expert systems of the 1970s, recent advances in machine ...

07/26/2018

The Secret Formula for Choosing the Right Next Role:
The best careers are not defined by titles or resume bullet points.

- Kate Matsudaira

Changing jobs—especially the higher up you get in your career—is a complex process. There are so many factors to consider, and often the factors that stand out most are the ones that matter the least: fancy titles, exciting projects, tempting promises of future success

But those factors that seem so valuable in the moment are just that—they are momentary. Your career isn't just about this one next step you're taking. Your career is a journey that will last a long time.

It is smarter to invest in your long-term success. Focus on factors that will increase your career capital and make you a more valuable hire in your next role, and the one after that, and the one after that.

When you are looking at the options for your next role, there are smarter choices that you can make. Here are the most important factors to consider when picking your next opportunity.

Changing jobs—especially the higher up you get in your career—is a complex process. There are so many factors to consider, and often the factors that stand out most are the ones that matter the least: fancy titles, exciting projects, tempting promises of future success

07/18/2018

The Mythos of Model Interpretability

In machine learning, the concept of interpretability is both important and slippery.

- Zachary C. Lipton

Supervised machine-learning models boast remarkable predictive capabilities. But can you trust your model? Will it work in deployment? What else can it tell you about the world? Models should be not only good, but also interpretable, yet the task of interpretation appears underspecified. The academic literature has provided diverse and sometimes non-overlapping motivations for interpretability and has offered myriad techniques for rendering interpretable models. Despite this ambiguity, many authors proclaim their models to be interpretable axiomatically, absent further argument. Problematically, it is not clear what common properties unite these techniques.

This article seeks to refine the discourse on interpretability. First it examines the objectives of previous papers addressing interpretability, finding them to be diverse and occasionally discordant. Then, it explores model properties and techniques thought to confer interpretability, identifying transparency to humans and post hoc explanations as competing concepts. Throughout, the feasibility and desirability of different notions of interpretability are discussed. The article questions the oft-made assertions that linear models are interpretable and that deep neural networks are not.

Supervised machine-learning models boast remarkable predictive capabilities. But can you trust your model? Will it work in deployment? What else can it tell you about the world? Models should be not only good, but also interpretable, yet the task of interpretation appears underspecified. The academi...

07/11/2018

Everything Sysadmin

GitOps: A Path to More Self-service IT
(IaC + PR = GitOps)

- Thomas A. Limoncelli

GitOps lowers the cost of creating self-service IT systems, enabling self-service operations where previously they could not be justified. It improves the ability to operate the system safely, permitting regular users to make big changes. Safety improves as more tests are added. Security audits become easier as every change is tracked.

You've written a new web application and would like it to be added to your organization's web load balancer. The load balancer is complex; its configuration is maintained by highly trained experts on the network operations team. You file a ticket with the team, wait, wait, wait, have a discussion wi...