Quality assurance · Production

WEX builds an agentic SDLC with Augment Code, compressing a seven-to-nine-month refactor into five days

The problem

WEX's SDLC relied on manual context-gathering that required scheduling breakout sessions with lead architects, leaving partner requests and secondary backlog items chronically deprioritized as engineers focused on core product work across very large, complex codebases.

Workflow diagram · grounded in source
1
Codebase query trigger
trigger
“Bill opened the codebase and ran a prompt”
2
AI synthesizes codebase context
ai_action
“What came back was a plain-English description of exactly how the process worked — its objectives, the workflow, and what it was accomplishing”
3
Product intake agent
ai_action
“An agent breaks problem statements down by domain mapping, identifying which teams own which capabilities, defining accountability boundaries, and preventing scope sprawl. It outputs epics, stories, and tasks formatted to a standard that…”
4
AI code generation
ai_action
“Engineers open VS Code, reference a ticket, and Augment handles code generation. The context pipeline gives Augment what it needs to write with architectural awareness, not just local file context.”
5
Code review agent
validation
“The code review agent is live and being evaluated. WEX is tracking comment acceptance rates, types of issues caught, and noise reduction. It gives engineers a first pass that cleans up the obvious issues before a teammate sees the PR. Wh…”
6
Security and QA gate
validation
“An InfoSec agent running static and dynamic application security testing checks automatically on every PR, feeding into a QA agent that validates expected behavior”
7
Human intent review
human_review
“with humans reviewing direction and intent rather than implementation detail”
Reported outcome

WEX compressed a seven-to-nine-month webforms refactor into five days, resolves complex partner requests in a single afternoon, and generates bug fixes with full test coverage in minutes — with product managers now able to navigate sophisticated system flows independently.

Reported metrics
Webforms refactor — original estimateseven-to-nine-month
Webforms refactor — actual completion timefive days
Refactor time reduction (summary)multi-month refactor in less than a week
Partner request resolution timea single afternoon
Show all 5 reported metrics
webforms refactor — original estimateseven-to-nine-month
webforms refactor — actual completion timefive days
refactor time reduction (summary)multi-month refactor in less than a week
partner request resolution timea single afternoon
bug fix with test coveragefull test coverage in minutes
Reported stack
AugmentVS CodeSnyk
Source
https://www.augmentcode.com/customers/WEX_automating-SDLC
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

WEX compressed a seven-to-nine-month webforms refactor into five days, resolves complex partner requests in a single afternoon, and generates bug fixes with full test coverage in minutes — with product managers now ab…

What tools did this team use?

Augment, VS Code, Snyk.

What results were reported?

Webforms refactor — original estimate: seven-to-nine-month; Webforms refactor — actual completion time: five days; Refactor time reduction (summary): multi-month refactor in less than a week; Partner request resolution time: a single afternoon (source-reported, not independently verified).

How is this quality assurance AI workflow structured?

Codebase query trigger → AI synthesizes codebase context → Product intake agent → AI code generation → Code review agent → Security and QA gate → Human intent review.

Related quality assurance cases