GE Aerospace
Devin coding agent for user stories, development and testing
GE Aerospace reported almost twice the user-story output on one team using Devin
Problem
Customer feature requests competed for limited engineering capacity. Translating complex features into development work also consumed time.
Approach
GE Aerospace began with a small engineering group and test cases that included legacy-code modernization. It expanded use gradually. Devin generated user stories from the codebase and carried work through development and testing, with engineers directing and reviewing it.
Results
Takeaway
The reported increase concerned one constrained team over a short period. GE Aerospace did not disclose a company-wide productivity result.
Categories
- Size
- Large enterprise
- Industry
- Manufacturing
- Business function
- IT and R&D
- Innovation type
- Process innovation
- AI technology
- AI agentsGenerative AI
- Adoption
- Vendor solution
- Evidence
- Vendor case study
Sources
Source summaries
Cognition described GE Aerospace's gradual deployment of Devin across software development tasks. It quoted an executive reporting almost twice the user-story output capacity for one team over 3 months.
Full details are in the original sources.