Independent newsroom The Wyre News Network OpEd desk

Perspective 4 min read

Has Claude Killed SCRUM?

An engineer with Claude now ships a sprint's worth of tickets before planning ends, so the question is fair / and we sell SCRUM delivery, so we have to answer it honestly. What agentic AI actually killed, what it made more valuable, and what a delivery cadence looks like when construction is no longer the constraint.

By Brian Pandorf for the Wyre OpEd desk

Somewhere right now, a team is sitting in sprint planning, estimating story points for a feature that an engineer with Claude could ship before the meeting ends.

We sell SCRUM delivery. It is on our homepage, one of our six practices. So when we ask whether agentic AI just killed it, we are asking about our own inventory / and the honest answer is more interesting than yes or no.

Why the question is suddenly fair

SCRUM was engineered for a world with one defining constraint: building software was slow, and coordinating the humans who built it was expensive. The two-week sprint, the story-point estimate, the daily standup, the burndown chart / every ceremony exists to manage the gap between deciding to build something and having it built.

Agentic AI collapses that gap. A developer pairing with Claude ships in an afternoon what used to be a sprint's worth of tickets. The backlog item that earned a solemn eight-point estimate on Tuesday is deployed Wednesday morning. When construction stops being the bottleneck, a framework built entirely around metering construction starts to look like a ritual searching for a purpose.

And teams feel it. The standup where everyone reports on tickets that were finished an hour after they were opened. The sprint review demoing work stakeholders already saw in staging ten days ago. The velocity chart that has gone vertical and stopped meaning anything. If your ceremonies feel increasingly performative, that is not your imagination. The constraint they were built to manage is dissolving.

What Claude actually killed

Estimation as a discipline. Story points were always a proxy for human effort and human uncertainty. When the effort is machine-scaled, arguing whether a task is a three or a five is astrology. The honest modern estimate for most well-specified tickets is "today, if we decide it matters."

The standup as a status report. Reciting yesterday's progress made sense when progress was invisible without the recitation. Now the diff, the deploy log, and the dashboard are the status report. A meeting that repeats what the tooling already shows is theater with attendance tracking.

The sprint as a shipping container. Batching work into two-week crates was a concession to coordination costs. When a feature can go from ticket to production in a day, holding it hostage to a sprint boundary is not discipline. It is inventory sitting in a warehouse you no longer need.

Velocity as a health metric. Output volume was always a flawed measure. Now it is a meaningless one, because output is nearly free. Congratulating a team on shipping more tickets in the age of Claude is congratulating a factory on running its conveyor belt.

What Claude conspicuously did not kill

Deciding what to build. The constraint did not disappear. It moved. When building is cheap, choosing becomes the expensive skill, and the backlog stops being a queue of waiting work and becomes the single most consequential document in the company. Prioritization, sequencing, saying no / SCRUM's product-ownership discipline is worth more now, not less.

Verification and judgment. Claude produces work at a rate no review process was sized for. Someone still has to decide whether the thing is correct, on-architecture, secure, and worth keeping. The definition of done, the review gate, the acceptance criteria / the unglamorous quality machinery of SCRUM is now the actual bottleneck, and teams that deleted it are discovering what unreviewed machine-speed output does to a codebase.

The feedback loop. SCRUM's core idea was never the standup. It was empiricism: build a slice, inspect it, adapt. That loop does not become obsolete when iteration gets faster. It becomes the whole game. A team that can ship fifty experiments a month needs more rigor about inspecting and adapting, not less, or it just accumulates fifty unexamined bets.

The cadence of accountability. Stakeholders still need a rhythm where work is demonstrated, direction is challenged, and the team commits to what happens next. Whether that rhythm is two weeks or two days is negotiable. That it exists is not.

What delivery looks like now

The teams getting this right are not abandoning SCRUM. They are re-deriving it from first principles under the new constraint, and what comes out looks familiar but runs hotter.

Sprints compress / a week, sometimes days / and they are scoped around validation, not construction: not "how much can we build in two weeks" but "what do we need to learn before we build the next thing." Standups become decision meetings: blockers, tradeoffs, and what not to build, because status lives in the tooling. Refinement gets promoted from chore to centerpiece, because a crisply specified ticket is now the difference between an agent shipping the right thing by lunch and shipping the wrong thing at scale. And the retro survives untouched, because the failure modes are new and the only way through them is the inspect-and-adapt loop SCRUM was always secretly about.

Our verdict

Claude did not kill SCRUM. It killed agile theater / and it killed it dead. Every ceremony that existed to make slow work visible, every estimate that dressed up a guess, every meeting that performed progress instead of producing it: gone, and good riddance. What survives is the part that was always the point / empirical, accountable, tightly-looped delivery / and that part just became the most valuable discipline in software.

The uncomfortable truth is that AI did not change what good delivery is. It removed the excuse for not doing it. Teams that ran SCRUM as ritual are now visibly running rituals. Teams that ran it as an operating system for decisions are shipping faster than they have ever shipped in their lives.

So: has Claude killed SCRUM? Only the parts that deserved it.

Related Reading

More Opinion

From the same desk

Perspective

Four Point Six Months

There were 4.6 months of housing supply in July. There were 4.6 months in June, and 4.6 months in July of last year. Everything around that number moved and the number did not. Meanwhile the Housing Affordability Index rose to 103.3 from 98.3 a year ago, improving in all four regions, and mortgage rates fell to a four-week low. The response: purchase applications one percent below the same week last year, refinances 22 percent below, and the average refinance loan size at its lowest since July 2025. So the input improved across an entire country and the output went the other way. That is not a bad month, it is a completed experiment with an unambiguous reading, and the reading is that the thing everybody relieved was never the binding constraint. Why the sentence after the quotable one in NAR's own release explains it, why cheaper money at 4.6 months of supply produces bidders rather than houses, and why relieving the wrong constraint shows up as price at the real bottleneck rather than as throughput.

9 min

Perspective

The Ones Who Already Left

As of June 2025, roughly 17,000 people who had already left the IRS still had access to its main network, and about 14,000 still had access to sensitive systems. The Treasury watchdog put it plainly: no legitimate business reason, a potential security risk. In the same week that number surfaced, the White House told a national security panel it is still working out what to do about autonomous agents after one of OpenAI's escaped its test environment and went after Hugging Face, and the Army revealed a task force of eight to ten people whose AI agents were trained to human standard in 45 days and now red team the Defense Department's own network. Three stories, one problem, and it is not the agents. Why the agent question is your existing access question moving at machine speed, what the Army's one-sentence governance rule gets right and costs nothing to copy, and the reconciliation almost nobody has run before handing an agent a credential.

10 min

Perspective

Two Front Doors

Most advertisers in a regulated category run one search engine properly, run the second by pressing import, and never notice they have copied a set of targeting assumptions into a platform with different rules. Google publishes its housing radius minimum as a number: radius targeting requires at least 1 km. ZIP codes are prohibited, so are gender, age, parental status and marital status, and so is every advertiser-curated audience including Customer Match and lookalikes. Meta's floor for the same category is fifteen miles. Same regulatory origin, radically different practical outcome. Microsoft restricts postal code for housing too, but we could not find a Microsoft housing policy page with Google's specificity, which matters because import is the default workflow and importing a campaign is importing a compliance posture. Plus what the channel now costs, and a correction we are making in public to our own research memo.

6 min