AI-Assisted ERP Development for NetSuite SaaS Platform
Cleveroad embedded an AI-assisted team into Proprio Cloud Solutions to deliver the Orion platform and a Field Service mobile app while replacing manual QA with full test automation across three client environments.
Industry
ERP
Team
5 members
Launched
10 months
Country
USA
35%
higher sprint output with AI-assisted delivery
4
major Orion platform releases shipped on schedule
3
client environments with full automated regression
About our Client
Proprio Cloud Solutions is a Michigan-based SaaS company building NetSuite-integrated platforms for the contract furniture industry. Its flagship platform, Orion, runs procurement, quoting, and operational workflows for MillerKnoll-affiliated dealers and the broader dealer network — a sector that has relied on decades-old systems for far too long. Proprio is modernizing that ecosystem from inside NetSuite.
Client’s Business Goals
Proprio was running two product streams in parallel under fixed delivery deadlines and a constrained internal team. The Orion platform required ongoing release work across three dealer environments. A new Field Service mobile app needed to reach MVP with offline-first capability and live NetSuite synchronization. And manual regression testing was eating three full working days per release — one day per environment, multiplied across three. As the platform grew, that cost scaled with it.
The company needed a development partner to:
Accelerate Orion platform delivery by standardizing workflows and building reusable components across multiple client environments. Target: four major releases shipped on schedule.
Deliver a Field Service mobile app MVP supporting work order management, crew tracking, punch lists, offline functionality, and real-time synchronization with NetSuite.
Eliminate manual regression testing as a release bottleneck by establishing automated end-to-end test coverage across all three client environments.
Why This Couldn't Be Solved Internally
The staffing requirement was specific: a senior full-stack engineer with real NetSuite experience (SuiteScript, SuiteFlow, SuiteQL) who could also work across a React Native codebase, a manual QA engineer, an automation QA engineer, a business analyst, and a project manager. Five specialists, with one working inside a niche ERP ecosystem uncommon outside the NetSuite domain. Finding that combination in-house wasn't realistic within Proprio's timeline. The same shape applies to most SaaS firms running NetSuite verticals — constrained internal capacity, fixed delivery dates, and a niche skill profile that's hard to source on short notice.
Cooperation Process and Outcomes
The engagement ran for 10 months across two parallel workstreams, structured around a rapid ramp-up phase followed by continuous delivery.
Development in Detail
The engagement was structured as two parallel workstreams under continuous delivery: ongoing Orion platform releases on the NetSuite SuiteApp, and the build of the Field Service mobile app from spec to MVP. End-to-end QA automation ran across both, integrated into Proprio's CI/CD pipeline from day one.
- Orion Platform. Cleveroad’s full-stack engineer worked inside Proprio’s core NetSuite SuiteApp, picking up SuiteScript, SuiteFlow, and SuiteQL within two weeks using Claude Code to trace execution across modules. The team shipped four major platform versions on schedule, built reusable SuiteScript components deployable across different dealer configurations, and maintained consistency across multi-subsidiary models.
- Field Service Mobile App. Together with Proprio’s internal team, Cleveroad delivered the MVP of a Field Service mobile application in React Native. The app supports on-site operations with an offline-first architecture and background data sync for locations with unreliable connectivity. Key capabilities: work order management, crew tracking, punch lists, and real-time synchronization with the Orion/NetSuite backend.
- QA Automation. Cleveroad’s QA engineers stood up end-to-end automation using Playwright, integrated into Proprio’s CI/CD pipeline. The team replaced manual regression entirely: full automated coverage across three client environments, each serving a separate dealer configuration. Playwright test scripts were generated fusing AI-assisted tooling, cutting script creation time from a full day to ≈two hours.
Over 10 months, Cleveroad's team shipped four major Orion platform versions on schedule across three client environments. The Field Service mobile app reached MVP and is running in a controlled pilot with 12 Proprio dealer technicians across 33 dealer locations. Pilot data so far: ~280 work orders synced from the field, 98% offline sync success rate over unreliable connectivity. Automated regression replaced the manual process entirely, running on every deployment through the CI/CD pipeline. Zero missed deadlines, zero scope reductions across the full engagement. The engagement is structured to support Proprio's path toward SOC 2 readiness.
The Role of AI-Assisted Development in Product Delivery
Cleveroad embedded an AI-assisted team into Proprio's development workflow on a staff-augmentation model: a senior full-stack engineer (NetSuite + React Native), a manual QA engineer, an automation QA engineer, a business analyst, and a project manager — five specialists working as one unit. Every role used Claude Code inside a defined workflow.
The engineer parsed the Orion codebase to trace SuiteScript execution flows and generate pull requests with tests, onboarding to SuiteScript, SuiteFlow, and SuiteQL in two weeks despite no prior NetSuite background. QA engineers generated Playwright scripts directly from acceptance criteria, dropping script creation from a full working day to roughly two hours per scenario. The project manager decomposed features into sprint-ready stories and prepared sprint summaries that previously took 2–3 hours per sprint in under ten minutes.
A human reviewed every AI-generated output before it shipped: code went through standard PR review, test scripts were validated against test data, and acceptance criteria were signed off by the BA and PM before stories went into a sprint. AI was allowed to draft, refactor, generate tests, and accelerate documentation. AI was not allowed to touch security-sensitive code paths without two-engineer review, modify production data, or sign off its own work.
Sprint output rose 30–40% against a conventionally staffed team of the same size, measured by Proprio against story points completed per sprint versus baseline velocity from previous cycles. Defect rates stayed flat against Proprio's pre-engagement baseline through the same period.
Product Essentials
The engagement was delivered across two products, each operating as a distinct workstream under Cleveroad's project management.
Orion Platform
Orion Platform
Orion is a NetSuite SuiteApp that runs procurement, quoting, and operational workflows for contract furniture dealers, including MillerKnoll-affiliated dealers operating at enterprise scale. Each dealer organization operates in an isolated tenant environment with its own data, configurations, and approval logic.
The platform handles multi-subsidiary data models where purchase orders, quotes, and operational records stay consistent across entity boundaries. Reusable components adapt to different dealer configurations without per-client customization.
Access control runs through NetSuite's native security model — role-based access via NetSuite role configurations, with audit logging of procurement and quoting actions handled by NetSuite's standard activity tracking.
Field Service App
Field Service App
A cross-platform mobile application for field technicians managing on-site installations and service calls. The app covers work order management, crew assignment and tracking, and punch list completion.
It runs on an offline-first architecture with background data sync, so technicians in warehouses or job sites with poor connectivity can keep working. All field data synchronizes back to the Orion/NetSuite backend in real time once the connection is restored.
The app uses NetSuite-issued authentication tokens — no credentials stored on device, all session data encrypted at rest. The app is currently deployed to a controlled pilot group of Proprio dealer technicians for field validation.
Technology stack
ERP Platform
Oracle NetSuite
SuiteScript
SuiteFlow
SuiteQL
Cross-platform
React Native
QA
Playwright
AI Tooling
Claude Code
Value for Our Client
30–40% higher sprint output using AI
Sprint output increased 30–40% compared to a conventionally staffed team of the same size. Proprio measured the difference by comparing story points completed per sprint against baseline velocity from previous development cycles. The AI-assisted model delivered higher throughput per person per sprint at the cost of the original team size. No additional spend.
Manual regression eliminated
Automated end-to-end QA now runs across all three client environments via a CI/CD pipeline for web and mobile. Regression testing, which previously took three days per release, is no longer a bottleneck. Sprint summaries, once requiring 2–3 hours, are now generated in under ten minutes using a Claude Code workflow that compiles engineering and QA data.
On-time delivery across both products
Four major Orion platform releases shipped on schedule. Field Service mobile app MVP delivered with work order management, crew tracking, punch lists, offline functionality, and real-time NetSuite sync. Currently in a controlled dealer pilot. No missed deadlines or scope reductions across the 10-month engagement.
Our Clients Say
USACTO at Proprio Cloud Solutions
“Sprint velocity went up 30 to 40 percent with Cleveroad's AI-assisted team, without growing headcount or dropping code quality.”