SHIFT Inc. (“SHIFT” or the “Company”), a provider of software quality assurance (QA) and testing services, today announced the launch of its AI Test Agent, a solution that automates QA processes starting from software specifications and operates autonomously 24 hours a day, 365 days a year. The AI Test Agent enables early identification of quality risks and is expected to reduce test execution time by up to 80%, supporting faster and more reliable software releases.

The AI Test Agent continuously performs test design and execution based on system specifications and design documents- tasks that have traditionally relied heavily on manual effort. By leveraging SHIFT’s proprietary quality assurance framework, extensive historical QA data, and the expertise of experienced test engineers as AI context, the solution executes tests autonomously and efficiently while maintaining accuracy. This allows quality issues to be identified earlier in the development lifecycle without slowing development speed.

Test execution results are accumulated in SHIFT’s integrated software test management platform, CAT, and are continuously incorporated into AI learning processes. Through repeated execution and learning, the AI Test Agent improves test design quality over time, creating a self‑evolving QA cycle that enhances both efficiency and precision.

Launch of the AI Test Center “Nemuranai”

In conjunction with the launch of the AI Test Agent, SHIFT has established an AI‑powered test center named “Nemuranai” (meaning “never sleeping” in Japanese). At Nemuranai, test engineers serve as AI Test Managers, supervising AI Test Agents rather than executing tests manually. The center is designed to support parallel test execution across up to 10 environments simultaneously, enabling rapid visibility into quality risks and development issues.
Service website:　https://contents.shiftinc.jp/nemuranai/　(Japanese Only)

Expected Impact of AI Test Agent–Driven Testing

In a validation study conducted on a portal website with approximately 1,000 test cases, SHIFT estimates that the use of the AI Test Agent can reduce test execution time by up to 80% compared to conventional testing approaches. While actual results may vary depending on system characteristics, scale, and execution requirements, SHIFT expects similar benefits to be achievable in larger and more complex projects.

The solution is particularly well suited for development environments with rapid release cycles and frequent specification changes, where accelerating testing cycles while maintaining quality is critical.

Overview of the AI Test Agent

The AI Test Agent operates autonomously on a 24/7/365 basis, continuously detecting software quality issues. It supports parallel execution across up to 10 environments, enabling timely and actionable visualization of quality risks. SHIFT plans to progressively deploy the AI Test Agent across quality assurance projects in a wide range of industries, including critical infrastructure sectors.

Key Capabilities
・AI‑Executable Test Design from Existing Documentation
Existing system specifications and design documents are converted into test design formats optimized for AI interpretation and execution, helping prevent gaps in test coverage and maintain high testing accuracy.

・Integration of SHIFT’s Accumulated Quality Knowledge
SHIFT has accumulated extensive QA data through participation in software testing projects across industries, including more than two million defect records, along with test cases, testing methodologies, test perspectives, and industry‑specific quality factors. This knowledge, combined with insights from experienced test engineers, is embedded into AI contexts to replicate SHIFT’s quality assurance methodology.

・Automated and Continuous Parallel Test Execution
By automating QA operations, the AI Test Agent enables autonomous, continuous, and parallel test execution at a scale not feasible through manual processes, significantly shortening testing cycles and improving test coverage.

・Self‑Evolving Improvement of Test Design Quality
Test execution results are stored in the CAT platform and continuously learned by AI, enabling ongoing improvement in test design quality as execution volume increases.

—
Forward-Looking Statements

This press release contains forward‑looking statements within the meaning of the Private Securities Litigation Reform Act of 1995, including statements regarding expected benefits, performance, scalability, and future adoption of the Company’s AI Test Agent and AI‑driven QA services. These statements are based on current expectations and assumptions and involve risks and uncertainties that could cause actual results to differ materially. Investors are cautioned not to place undue reliance on these forward‑looking statements.

Investor Relations Contact

Investor Relations
SHIFT Inc.
Email: ir@shiftinc.jp