TL;DR
Get the latest gadgets delivered free — and shop member deals
- Fast, free delivery on millions of items
- Access to Prime Big Deal Days deals on October 6–7
- Prime Video, Amazon Music and more included
A report by The Pragmatic Engineer says AI coding tools are changing how software engineers work in 2026, with some experienced developers managing several agents in parallel rather than writing code line by line. The report also flags concerns about code quality and reviews, while stressing that teams and planning still matter; it does not establish industry-wide adoption rates.
AI coding agents are changing how some software engineers work, according to a 2026 industry snapshot from The Pragmatic Engineer, which reports that experienced developers increasingly run several agents in parallel instead of writing code by hand. The account, based on interviews, visits to technology companies and data shared by several firms, also warns that code quality and reliability have weakened and that some code reviews have become performative. It is a reported snapshot, not a measured survey of the entire industry.
The report’s author presented the findings at LDX3, an engineering leadership conference in New York attended by more than 2,000 people. The research drew on visits to AI labs OpenAI and Anthropic, conversations with startups and technology companies, and unpublished data from GitHub, Factory AI and Linear. The source does not provide enough detail about those datasets in the supplied material to assess their methods or representativeness.
One clear theme is the move from single-task coding toward multiple concurrent agent sessions. Claude Code creator Boris Cherny said he uses five terminal sessions, each with a separate repository checkout, alongside five to 10 Claude sessions on the web. Linear software engineer Dima Zaytsev described keeping five to 10 local worktrees and moving between agents as they produce work. These accounts illustrate individual practices; they do not prove that most engineers work this way.
The report also describes a shift in the engineer’s role, as coding agents generate more of the implementation and people coordinate tasks, test output and review results. At the same time, it says assumptions about code output have broken down: reviews may no longer provide meaningful scrutiny, and quality and reliability are concerns. The supplied source does not quantify these problems or specify how widely they occur.
How AI Changes Engineering Work
The shift matters because software development depends not only on how quickly code is produced, but also on whether teams can understand, test and maintain it. If engineers spend more time directing agents and checking their output, companies may need to change review processes, testing practices and measures of productivity. Faster code generation alone does not establish that products are delivered faster or work more reliably.
The report’s tension is practical: agents can let developers handle more work at once, but parallel output also creates more material to validate. If reviews become a formality, defects may be harder to catch. The account raises questions for engineering leaders about accountability and quality controls, while offering no industry-wide figures on productivity gains, errors or costs.
As an affiliate, we earn on qualifying purchases.
From Coding Tools to Agent Workflows
Software teams have adapted to major changes before, including the internet, smartphones, cloud computing and new programming languages and frameworks. The report argues that the current AI shift is happening at a different pace and scale. Martin Fowler, a software engineering expert quoted by The Pragmatic Engineer, said AI’s impact was larger than earlier changes he had experienced.
The report links the acceleration to improvements in coding models late in 2025 and describes practices seen in 2026. Its author says many engineers appear to have reduced or stopped hand-writing code, but the source provides no survey establishing how many. It also notes that teams and planning remain important and says non-engineers are not broadly shipping code, complicating the idea that AI simply removes the need for software teams.
“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 expert, as quoted in The Pragmatic Engineer
Adoption and Quality Still Unmeasured
The report does not establish how common multi-agent workflows are across software teams, or whether they improve delivery speed, cost or outcomes. The examples come from experienced engineers and technology companies, and the source material does not include methods or sample sizes for the unpublished data it references.
It also remains unclear how large the reported quality and reliability problems are, which kinds of projects are most affected, and whether organizations are changing review and testing practices to address them. The report’s observations indicate concerns, not quantified industry-wide conclusions.
Companies Test New Agent Systems
The report expects cloud-based coding agents and supporting software infrastructure to develop further, alongside new practices for coordinating and checking agent output. These are forecasts in the report, not confirmed outcomes or a published timetable.
For now, the next test is whether teams can turn faster code generation into dependable software. Engineering leaders will need clearer evidence on agent performance, review effectiveness and reliability before the individual workflows described in the report can be treated as standard practice. The source does not identify a specific upcoming milestone or release date.
Key Questions
What is changing in software engineering in 2026?
The Pragmatic Engineer reports that some developers increasingly use multiple AI coding agents at once, shifting part of their work from writing code toward directing and checking agent output.
Does the report show that most engineers have stopped writing code by hand?
No. The report says there are signs of that shift, but the supplied material gives no representative survey or adoption rate to establish how widespread it is.
What concerns does the report raise?
It points to concerns about code quality, reliability and reviews that may not provide meaningful scrutiny. The source does not quantify these problems.
Do AI coding agents mean software teams are no longer needed?
The report does not make that conclusion. It says teams and planning remain important, even as coding tools and working practices change.
Source: rss
Halloween Picks
halloween
As an affiliate, we earn on qualifying purchases.
