TL;DR
Get business pricing on monitors, keyboards and dev gear
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
A report based on a keynote to more than 2,000 engineering leaders describes AI as driving rapid changes in software development, including engineers coordinating multiple coding agents instead of writing code by hand. The report also flags weaker reliability and code reviews that may not adequately check AI-generated work; the scale of those problems and how broadly the trends apply are not quantified.
A report on the state of the tech industry in 2026 says AI coding tools are changing how software engineers work, with some experienced developers managing five to 10 agent sessions at once rather than writing each line of code themselves. The report, based on a keynote at the LDX3 engineering leadership conference in New York, also identifies concerns about code quality, reliability and reviews as AI-generated code becomes more common.
The Pragmatic Engineer report draws on the author’s conversations with teams at AI labs including OpenAI and Anthropic, as well as startups and technology companies, and on data described as unpublished from GitHub, Factory AI and Linear. The source does not provide the underlying datasets or a method for measuring how widespread the reported practices are, so its observations are best read as a snapshot from industry interviews and conference research, not a representative survey of all software workers.
Several developers quoted in the report describe running coding agents in parallel across separate workspaces or terminal sessions. Boris Cherny, identified as the creator of Claude Code, said he uses five terminal tabs and runs five to 10 Claude sessions on Claude Web alongside local sessions. Linear software engineer Dima Zaytsev described rotating between five to 10 local worktrees, prompting one agent and checking another’s output while it works.
The report says the changes extend beyond individual workflows: the role of the IDE is fading, and expectations about how much code engineers write by hand are shifting. At the same time, it argues that familiar practices have not disappeared: teams and planning still matter, and it says non-engineers are not yet broadly shipping code. It also warns that code reviews can become “theatrical” and that quality and reliability have fallen, though it does not include metrics establishing the size or cause of those problems.
AI Changes the Engineering Workflow
The shift matters because software development is not only changing at the point where code is written. When engineers supervise several agents at once, teams may need to rethink how they divide work, check changes and decide who is accountable for failures. The report’s account suggests that coordination and verification are taking a larger role alongside coding itself.
That creates a practical tension for companies adopting these tools. Agents can produce code while a developer works on another task, but parallel output still needs to be tested and reviewed. If reviews are treated as a formality, the report’s concerns about reliability and quality could affect products and the engineers responsible for maintaining them. The source describes this risk but does not establish a measured industry-wide decline.
As an affiliate, we earn on qualifying purchases.
From Coding Tools to Agent Work
The report places the current changes against earlier shifts familiar to technology workers, including the spread of the internet, smartphones, cloud computing and newer programming languages and frameworks. Its argument is that AI is changing work at a greater scale and pace than those earlier developments. That is an assessment from the report’s author and interviewees, rather than a quantified comparison across technology eras.
It points to stronger coding models near the end of 2025 as a recent catalyst for the shift described in 2026. AI-assisted programming had already prompted debate about whether software engineers would continue writing code by hand. The newer examples in the report show some developers using agents as parallel collaborators, while retaining responsibility for selecting tasks and checking results.
“Nothing has hit with the magnitude of AI. This is a whole size difference from anything that we’ve faced before.”
— Martin Fowler, software engineering author and industry veteran, quoted from The Pragmatic Summit
integrated development environment (IDE) with AI support
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
How Broad Are These Changes?
The report does not establish what share of engineers now rely mainly on agents, or how typical the quoted developers are. Its claims about fewer engineers writing code by hand, a fading IDE and declining quality are not accompanied in the supplied material by industry-wide figures, defined time periods or comparison baselines.
It is also unclear how much of the reported quality and reliability concern is caused by AI-generated code, changes in review practices, or other factors. The source describes code reviews as “theatrical” but does not define a measurement for that term. The announced data from GitHub, Factory AI and Linear is mentioned without its findings or methodology in the supplied text.
code review software for AI-generated code
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
The Next Phase of Coding Agents
The report expects cloud-based coding agents and supporting infrastructure to develop further, alongside changes in how engineers supervise work and check software. Those are forward-looking expectations in the report, not confirmed outcomes or dated product announcements.
For engineering teams, the immediate test is whether parallel agents can fit into dependable development processes. Future evidence to watch includes published data on adoption, measurable results for software quality and reliability, and details of the tools and review practices companies put in place. The report does not give a specific timetable for those developments.
parallel coding agent management tools
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
What is changing in software engineering in 2026?
The report describes developers using AI coding agents to generate and review work in parallel, reducing how much code they write by hand. It presents these as observed trends, not as a quantified account of the whole industry.
How many AI agents are some developers using at once?
Developers quoted in the report describe working across roughly five to 10 sessions. Cherny described local terminal sessions and additional Claude sessions on the web; Zaytsev described switching between five to 10 local worktrees.
Does the report show that AI has reduced software quality?
No industry-wide measurement is provided in the supplied report material. The author flags lower quality and reliability as concerns, but the scale, causes and comparison baseline are not specified.
Are teams and planning becoming unnecessary?
No. The report says teams and planning remain important even as coding workflows change. It describes AI as altering development practices, not eliminating the need for coordination.
Source: rss
Halloween Picks
halloween
As an affiliate, we earn on qualifying purchases.
