Where do you start?
Four paths into RACE Programming, each matched to your current context. Pick the one that describes where your team is today.
What all four guides share
Every guide follows the same structure: Why (what breaks without RACE Programming in your context), What (what you're adopting), Who (which roles change and how), How (step-by-step, 90 days), and How to measure success (the metrics that confirm it's working).
Every guide also has the same organizer. The TPM (Technical Program Manager) runs the transition: the transition plan and the order in which teams move, the dependencies between converting teams and the ones still on the old process, the risk register, escalation, and the alignment of the sponsor and stakeholders around the Team Principal. The TPM owns the container, not the contents. AI Product keeps the product, the release plan and the Executable User Stories; the Forward Deployed Engineer keeps the architecture; the Team Principal keeps budget and priorities; line managers keep the people.
If you're new to RACE Programming, read the Framework overview first. If you're coming from Agile and want the conceptual bridge, read the Agentic Agile Manifesto.
From Scrum
Your team uses Scrum. Sprints, velocity, story points, a Scrum Master. You want to adopt RACE Programming without discarding what works.
Read the Scrum transition guide →Greenfield
Starting from scratch. No legacy constraints, no existing team rituals. Build the RACE Programming system right from day one.
Read the greenfield guide →Brownfield
An existing product, existing codebase. Adopting RACE Programming while maintaining live production without a hard cutover.
Read the brownfield guide →Staffing to client speed
Staff up when the client accelerates. Staff down when they slow. The accordion model maps EUS/week throughput to Pit Wall and Pit Crew allocations across Discovery mode and Realization mode.
Read the staffing guide →You already track these metrics. Here is how to read a rollout.
DORA, SPACE, DevEx, DX Core 4: a RACE Programming rollout moves the numbers you already watch, and changes what "good" means. One short guideline per framework.
Measuring with DORA
Commit to production-ready collapses almost by construction; deployment frequency and lead time to production follow your release cadence. The one that still tells the truth, and the clock DORA does not have.
Read the DORA guideline →Measuring with SPACE
Each of the five dimensions has a home in RACE Programming. The Activity dimension is the one naive instrumentation gets wrong.
Read the SPACE guideline →Measuring with DevEx
Feedback loops, cognitive load, flow state. The lens where a rollout shows up first, and most honestly.
Read the DevEx guideline →Measuring with DX Core 4
The unified scorecard. Its four dimensions map onto RACE Programming almost one to one, and it already balances speed the way the triangle does.
Read the DX Core 4 guideline →How RACE Programming compares
Side-by-side comparisons (roles, cadence, artifacts, feedback loops, and economics) against the frameworks teams use today: Scrum, SAFe, Kanban, Extreme Programming, Shape Up, DevOps, DORA, and Team Topologies.