vmblog logo 2024 wht (updated)

CircleCI Expands Smarter Testing to Power Independent Validation for AI-Driven Development

Share: 

David Marshall | Published: November 19, 2026

CircleCI announced new features available in its Smarter Testing suite. Smarter Testing helps engineering teams get faster feedback by running only the tests affected by a code change, and new Dynamic Parallelism adjusts the number of machines for each run to cut unnecessary compute costs, enabling organizations to maintain confidence in what they ship without sacrificing speed, efficiency, or control.

Testing can account for up to half of CI/CD pipeline time, leaving developers waiting for feedback or switching context while tests run. As AI coding agents accelerate code creation, engineering teams need ways to validate changes just as quickly, regardless of which tools or agents are writing the code.

Smarter Testing brings together several capabilities to help teams validate changes earlier, reduce unnecessary testing, and use compute more efficiently:

  • Check changes before pushing. Developers and AI coding agents can run affected tests locally, catching issues before code reaches CI at no additional charge. Agents can iterate faster without waiting on the pipeline.
  • Run only the tests that matter. Intelligent Test Selection uses test impact analysis to identify tests affected by a code change, so teams don’t have to run the full suite on every commit.
  • Use only the machines you need. Dynamic Parallelism automatically adjusts the number of machines based on the tests selected for each run, reducing unnecessary compute costs as volume grows.
  • Spend less time waiting on tests. Dynamic Test Splitting balances tests across machines to reduce idle time, while Auto Rerun Failed Tests retries only failed tests instead of rerunning the entire suite.

Early customers are already seeing the impact. Medallion, a healthcare technology company that automates provider credentialing and operations, has used Smarter Testing to manage growing development activity. With Dynamic Parallelism, the number of machines allocated to each test run adjusts automatically based on the work required, helping Medallion reduce unnecessary compute usage as its testing needs grow.

“With the rapid growth of AI-driven workflows, CircleCI’s Smarter Testing has helped us bring down our cost per pull request and reduce our build times, even as our pull request volume has gone up 10x and our test suite has grown 2.5x.”

William Yardley | Staff Infrastructure Engineer at Medallion

Kustomer, an AI-powered customer service platform, has also seen significant savings from running fewer unnecessary tests.

“Smarter Testing changed the economics of our CI. The cost of running our tests was cut by a quarter, and because only the tests affected by a change run, our engineers get feedback faster.”

Maria Volpe | Senior Software Engineer at Kustomer

Smarter Testing is built directly into CircleCI, with no additional vendor or infrastructure required. By reducing unnecessary test runs and helping developers and AI coding agents catch issues earlier, teams can improve the efficiency of the entire delivery process and lower the overall cost of getting a change into production.

“As AI coding agents make it easier to produce code, everything downstream needs to get smarter and faster, including how we validate it. Developers and agents should be able to check their work before pushing code, and a small change shouldn’t require the same tests and compute resources as a large one. Smarter Testing helps teams get feedback sooner and use their resources more efficiently without having to rethink their entire development process,” said Rob Zuber, CTO of CircleCI.