This is not just an internal tooling update. CVS frames the work as directly tied to healthcare outcomes: better developer experience means faster and safer delivery of products that touch millions of lives daily.
The Scale Problem: 10,000 Developers, 150,000 Repositories, 10+ CI/CD Tools
Jim Beyers, Vice President of Developer Experience at CVS Health, opened with the core challenge: CVS is operating at nation-spanning scale. Their business includes Oak Street Health clinics, Aetna, CVS Caremark, MinuteClinic, CVS Specialty, and thousands of pharmacy locations. They serve roughly 185 million people and manage millions of daily digital and physical interactions.
That operational size created a deeply complex engineering reality:
- Over 7 different source control systems
- More than 10 CI/CD tools
- An estimated 150,000 repositories
- Fragmented governance and inconsistent security controls
- A steep learning curve for onboarding and internal mobility
Mitch Hendel, an Executive Director at CVS Health, described the situation bluntly: developers were effectively “speaking different languages.” With so many toolchains in play, collaboration slowed, onboarding dragged, and platform teams spent most of their time just keeping the lights on. Even simple policy changes—like blocking an artifact with critical vulnerabilities—became painfully difficult to enforce across a fractured pipeline landscape.
The Vision: “Vibrancy” as a Technology and Culture Strategy
CVS Health is not positioning Repo Rally as a narrow migration project. Instead, it is part of a broader technology strategy they call Vibrancy.
Vibrancy, in CVS terms, means:
- Working with the speed and creativity of a startup
- While still benefiting from the scale, resources, and discipline of a Fortune 500 enterprise
- Empowering teams with autonomy
- Removing friction from delivery
- Enabling fast experimentation and iteration
Beyers emphasized Vibrancy as a mindset shift, not just a tech upgrade. The goal is to let engineers focus on real healthcare problems—rather than fighting legacy processes or tooling silos. Every reduction in friction, he argued, translates into better health outcomes, because it speeds up innovation pipelines that support patients, clinicians, and communities.
The Solution: Standardizing on GitHub EMU + GitHub Actions
To turn Vibrancy into operational reality, CVS chose GitHub EMU (Enterprise Managed Users) as its unified backbone. The standardization strategy aims to deliver four major improvements:
- One Source of Truth for Code
All repositories move into GitHub EMU, making code discoverable, searchable, and consistently governed.
- Unified CI/CD With GitHub Actions
Pipelines will be consolidated into one system so governance, security gating, and metrics become enforceable at scale.
- Day-One Developer Enablement
New engineers automatically get GitHub access without ticket queues or approvals, enabling immediate productivity.
- Reduced Cognitive Load
With one toolchain and one workflow path, developers don’t waste mental bandwidth navigating a complex tooling map.
Hendel also highlighted a subtle but powerful benefit: career mobility. In the old world, switching teams meant learning a new tool stack from scratch. Standardization removes that barrier, letting internal talent move more fluidly across the organization.
Repo Migration at Enterprise Scale: Irene’s Five-Phase Playbook
Irene, leading GitHub adoption and repo migrations, delivered the most tactical segment of the session. She described how CVS approached migration without even knowing the true size of its repo universe at the start.
Step 1: Inventory and Classification
They pooled data from seven source code management tools, creating a single spreadsheet containing 100,000+ repo entries. Each repo was evaluated for CI/CD tooling, ownership, and metadata, enabling CVS to group migrations into five phases.
Step 2: Building Automation for Speed
The team developed scripts capable of migrating 600 repositories in four hours, run by only two engineers. These scripts didn’t just transfer code—they also:
- repointed Jenkins pipelines to GitHub repos
- performed automated post-migration validation
- reduced manual clean-up effort
Step 3: Communication at Scale
Initial plans asked repo owners to choose migration dates. That failed due to conflicting priorities. CVS flipped the model:
- mandatory migration dates were assigned
- teams had to justify delays and commit to new dates
This enforced urgency while still allowing flexibility.
Unexpected Blockers and the Fixes CVS Built
Irene walked through the real friction points they encountered, and what they built to remove them:
1. Slow Progress
Problem: teams weren’t volunteering migration slots
Solution: enforced mandatory schedules with accountability
2. Corrupted Spreadsheet
Problem: 100,000-line sheet + many editors = constant corruption
Solution: built a dedicated migration app to manage repo data cleanly
3. Manual Emails
Problem: Irene personally emailing 600+ repo owners repeatedly
Solution: event-driven automated email workflows
- 7-day notices
- reminders
- completion confirmations with new repo URLs
4. Too Much Hands-On Script Work
Problem: daily migrations took 7+ hours with four engineers
Solution: upgraded scripts with
- automated triggers
- error handling
- automated validations
The pattern here is consistent with Vibrancy: whenever friction surfaced, CVS engineered it away.
Results So Far: 85% by End of 2025, Full Completion by 2027
The migration roadmap is aggressive and measurable:
- Q1–Q2 2025: discovery + planning, scripts validated → 15% complete
- Q3 2025: automation + forced scheduling → 50% complete
- Q4 2025 target: 85% complete
- Q1 2026: deprecate six SCM tools → 95% complete
- Q3 2026–Q1 2027: migrate remaining GitLab repos (with CI/CD conversion) → 100% complete
This is one of the largest repo consolidation efforts publicly described at GitHub Universe, especially given the regulated, high-risk environment of healthcare software.
What’s Next: Migrating Thousands of CI/CD Pipelines (With AI Help)
While repo migration is the first mountain, Hendel noted the harder climb is next: pipeline migration.
Thousands of pipelines exist with custom scripts and bespoke patterns. Doing “lift-and-shift” manually would dump heavy work on developers—the opposite of CVS’s DX mission.
So CVS is experimenting with:
- GitHub Copilot
- Copilot Agents
- POCs to automate pipeline conversion into GitHub Actions
The aspiration is bold: use AI to migrate pipelines so dev teams don’t have to.
Why This Matters Beyond CVS
Repo Rally is an internal engineering program, but its stakes go far beyond developer convenience.
For CVS, this transformation means:
- faster innovation in healthcare apps and services
- safer environments through enforceable security gating
- reduced operational spend on tool maintenance
- higher developer satisfaction and retention
- improved delivery reliability across a critical national health infrastructure
In other words, developer experience becomes patient experience downstream.
Final Takeaway
CVS Health’s Repo Rally shows what true enterprise-scale DX transformation looks like:
- not a one-off migration
- not a tooling “refresh”
- but a multi-year cultural and technical rebuild tied directly to mission outcomes
By unifying on GitHub EMU and GitHub Actions, automating everything that slows teams down, and preparing to use AI for the most complex next phase, CVS is building a developer platform designed for the next decade of healthcare innovation.
And if their guiding idea of Vibrancy holds true—startup speed with enterprise power—then this Repo Rally won’t just modernize CVS’s codebase. It may meaningfully modernize how healthcare technology is delivered for millions of people.
