Constructor Gears Quality Framework

Constructor Gears is evaluated across five quality vectors that guide
architecture, implementation, validation, and continuous improvement
throughout the software lifecycle.

Note: The North Star Metric is the primary indicator for each
quality vector. The Example Metrics column contains representative
metrics that teams may use to measure and monitor that vector. The
specific metrics can evolve over time.

Quality Vector Guiding Question Description North Star Metric Example Metrics (non-exhaustive)
Efficiency How quickly and economically can software be built, deployed, and operated? Minimize the time, effort, and cost required to design, build, deploy, and operate software through AI, automation, reuse, composability, extensibility, and operability. Total Cost of Ownership (TCO) Time from approved PRD to production; TCO to build and operate a feature, Gear, or product; Lead time for change; Cost per delivered feature; Infrastructure cost per transaction/workflow; Infrastructure cost per tenant/service
Reliability How dependable is it? Provide continuous, fault-tolerant service with predictable behavior during failures, upgrades, and peak load. Service availability (SLA) MTTR; MTBF; Failed workflow rate; Change failure rate; Successful deployment rate; Error rate; Disaster recovery success rate
Performance How fast does it execute? Deliver consistently low latency and high throughput while efficiently utilizing system resources. P99 workflow latency Transactions/workflows per second; Average response time; P99/P999 latency; Resource utilization (CPU/Memory) per transaction; % of performance SLAs met; Cold start time
Security How well is it protected? Protect applications, data, identities, and platform services through secure-by-design principles and continuous compliance. Critical security findings in production (Target: 0) Security policy compliance; Secrets management coverage; Mean time to remediate vulnerabilities; Dependency compliance; Secure coding compliance; Security incident rate
Versatility How many real-world scenarios can it support without building a new platform? Support the broadest range of business scenarios, industries, deployment models, and architectural patterns using a common platform. % of target business scenarios supported out of the box Supported business scenarios; Supported deployment models; Supported integration types; Supported business domains; Configuration vs. customization ratio; Feature coverage across target scenarios

Summary

The quality vectors are prioritized as follows:

  1. Efficiency — Minimize engineering effort and operational cost
    across the software lifecycle.
  2. Reliability — Ensure continuous, predictable operation.
  3. Performance — Deliver fast execution under production
    workloads.
  4. Security — Protect the platform by design and maintain
    continuous compliance.
  5. Versatility — Maximize the range of business scenarios
    supported by a common platform.