680 hours of manual work completed in 13 hours by automating tasks with Gemini CLI
Five days to build complex Java meta-framework instead of three months
KeyBank's pilot of Gemini CLI cut a three-month project to five days and saved 680+ hours of manual coding work.
KeyBank's pilot of Gemini CLI cut a three-month project to five days and saved 680+ hours of manual coding work.
How do you build on more than 200 years of banking expertise to power a digital-first future? For KeyBank, an American financial institution that operates approximately 950 branches and 1,200 ATMs across 15 states, technology has long been a strategic enabler, connecting relationship-driven banking with modern digital experiences. Recognizing that agility, security, and scale would define the next chapter of financial services, KeyBank set out to further strengthen its technology foundation to support faster innovation.
Starting in 2018, KeyBank shifted its data warehouse and fintech acquisitions to Google Cloud, a move that reduced costs and bolstered security while establishing a flexible foundation for growth. Yet, as the bank scaled its cloud presence, a new bottleneck emerged. While the infrastructure was modern, the millions of lines of code running on it often carried the weight of decades of technical debt.
This invisible friction meant that even the most talented developers were tethered to manual, repetitive tasks—like migrating legacy Java frameworks or conducting grueling, day-long code reviews—that slowed the pace of innovation. To truly unlock the power of the cloud, KeyBank needed a way to accelerate the development life cycle itself.
“We saw an opportunity to use generative AI to standardize coding processes and alleviate the mental load on developers,” says Joseph Steinbrenner, technical lead for KeyBank’s Card Technology Engineering team. Building on the bank’s existing foundation with Google Cloud, Steinbrenner began exploring a proof of concept (POC) with Gemini CLI to support development teams and empower them to focus more on creativity rather than routine maintenance.
We saw an opportunity to use generative AI to standardize coding processes and alleviate the mental load on developers.
Joseph Steinbrenner
Technical Lead, Card Technology Engineering, KeyBank
Accelerating delivery from months to days
To kick off the POC, KeyBank integrated Gemini CLI into daily developer workflows to tackle everything from routine documentation to high-stakes legacy migrations. As part of this effort, Steinbrenner used CLI to build a complex Java meta-framework to standardize migration paths for older systems. “This project previously would have taken roughly three months of manual effort,” he says. “With Gemini CLI, I completed the work in just five days.” Additionally, Gemini CLI reduced approximately 680 hours of manual work to 13 hours by automating routine tasks that previously drained developer resources.
Beyond raw speed, Gemini CLI improved the quality and clarity of the development process by serving as a first reviewer for the team. Previously, reviewing large code branches could take a full day, but Gemini now summarizes changes across branches, eliminating the noise and confusion of complex releases. Steinbrenner also uses the tool to translate technical complexity for the rest of the business. By scanning specific code areas, Gemini can generate high-level summaries that help teammates in non-technical roles understand how a function works before they head into meetings.
Gemini CLI has now become indispensable for daily development work. Steinbrenner uses it to analyze code sections and automatically generate standardized task formats for his developers, providing a clear overview of the work required without him having to write it from scratch. It’s also transformed how the team handles documentation, generating user guides and architectural diagrams in minutes. This automation extended to legacy code modernization: Gemini CLI helped migrate systems from JDK 8 to JDK 21 by processing thousands of lines of code and generating the necessary unit tests to ensure stability, a task that previously would’ve required an unquantifiable amount of manual labor.
This project previously would have taken roughly three months of manual effort. With Gemini CLI, I completed the work in just five days.
Joseph Steinbrenner
Technical Lead, Card Technology Engineering, KeyBank
Empowering the next generation of engineers
Gemini CLI has created a significant cultural shift within KeyBank’s engineering teams. By removing the manual grunt work, senior developers and tech leads have reclaimed hours of their day that were previously spent on repetitive tasks. This time is now being reinvested into high-value mentoring and training for junior engineers. “While AI can be a powerful assistant, it cannot replace the human element of engineering,” Steinbrenner says. “Instead, Gemini CLI empowers senior developers to help juniors become the next generation of solid senior developers.”
Looking ahead, KeyBank plans to expand the rollout of Gemini CLI across the enterprise. The goal is to use the tool’s massive context window and orchestration capabilities to continue modernizing legacy systems and bringing them into the modern era. For KeyBank, the success with Google Cloud is about so much more than simply writing code faster. It’s about building a more impactful, efficient, and collaborative engineering culture that can better serve its customers for years to come.
While AI can be a powerful assistant, it cannot replace the human element of engineering. Instead, Gemini CLI empowers senior developers to help juniors become the next generation of solid senior developers.
Joseph Steinbrenner
Technical Lead, Card Technology Engineering, KeyBank
KeyBank is one of the largest, bank-based financial services companies in the United States. Part of KeyCorp, it operates approximately 950 branches and 1,200 ATMs across 15 states.
Industry: Financial Services
Location: United States
Product: Gemini CLI
