Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2602.21568 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910032682024960 |
|---|---|
| author | Agrawal, Yuvraj Jain, Pallav |
| author_facet | Agrawal, Yuvraj Jain, Pallav |
| contents | Developer Productivity Dashboards are essential for visualizing DevOps performance metrics such as Deployment Frequency and Change Failure Rate (DORA). However, the utility of these dashboards is frequently undermined by data reliability issues. In early iterations of our platform, ad-hoc ingestion scripts (Cron jobs) led to "silent failures," where data gaps went undetected for days, eroding organizational trust. This paper reports on our experience migrating from legacy scheduling to a robust Extract-Load-Transform (ELT) pipeline using Directed Acyclic Graph (DAG) orchestration and Medallion Architecture. We detail the operational benefits of decoupling data extraction from transformation, the necessity of immutable raw history for metric redefinition, and the implementation of state-based dependency management. Our experience suggests that treating the metrics pipeline as a production-grade distributed system is a prerequisite for sustainable engineering analytics. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_21568 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | From Ad-Hoc Scripts to Orchestrated Pipelines: Architecting a Resilient ELT Framework for Developer Productivity Metrics Agrawal, Yuvraj Jain, Pallav Software Engineering Developer Productivity Dashboards are essential for visualizing DevOps performance metrics such as Deployment Frequency and Change Failure Rate (DORA). However, the utility of these dashboards is frequently undermined by data reliability issues. In early iterations of our platform, ad-hoc ingestion scripts (Cron jobs) led to "silent failures," where data gaps went undetected for days, eroding organizational trust. This paper reports on our experience migrating from legacy scheduling to a robust Extract-Load-Transform (ELT) pipeline using Directed Acyclic Graph (DAG) orchestration and Medallion Architecture. We detail the operational benefits of decoupling data extraction from transformation, the necessity of immutable raw history for metric redefinition, and the implementation of state-based dependency management. Our experience suggests that treating the metrics pipeline as a production-grade distributed system is a prerequisite for sustainable engineering analytics. |
| title | From Ad-Hoc Scripts to Orchestrated Pipelines: Architecting a Resilient ELT Framework for Developer Productivity Metrics |
| topic | Software Engineering |
| url | https://arxiv.org/abs/2602.21568 |