QC Automation Agent

The QC Automation Agent is a software system that automatically inspects drilling data quality across an entire portfolio of active wells. It checks 29 data modules for every well, computes weighted quality scores by operator, and publishes the results to a centralized tracking board. The agent replaces a manual inspection process that was time-consuming, inconsistent, and limited in how often it could run.

Why It Exists

Drilling operations generate large volumes of data across dozens of platform modules – assembly records, survey measurements, real-time sensor feeds, engineering plans, daily reports, and document uploads. Keeping that data complete and up to date is essential for operational safety, regulatory compliance, and decision-making.

Before automation, quality control was performed manually: a team member would open each well on the platform, navigate to each module, visually verify whether the expected data was present and current, and record the results. For a portfolio of approximately 115 active wells, this process took 6 to 7 hours per week and could only be performed once per week at that pace.

The manual approach had three fundamental limitations:

  1. Speed – 29 checks across 115 wells is over 3,300 individual inspections per cycle
  2. Consistency – two reviewers checking the same well could reach different conclusions depending on interpretation and fatigue
  3. Frequency – weekly checks meant data quality issues could persist for days before being caught

What the Agent Does

The agent performs the same 29 inspections, but it does so programmatically. It reads a list of wells from an input file, authenticates with the cloud platform, retrieves data from each module through the platform’s API, evaluates the data against a fixed set of rules, and publishes scores to a Monday.com board where account managers and leadership can review them.

Every check produces a clear result – YES (data is present and correct), NO (data is missing or incorrect), PARTIAL (partially complete), N/A (not applicable to this well), or INCONCLUSIVE (the agent could not determine the answer). The same data always produces the same result.

flowchart LR
    A["Well List\n(CSV Input)"] --> B["Authenticate\n& Connect"]
    B --> C["Inspect\n29 Modules"]
    C --> D["Score\nby Category"]
    D --> E["Publish to\nQC Board"]

    style A fill:#4a90d9,stroke:#2c5aa0,color:#fff
    style B fill:#5ba585,stroke:#3d7a5e,color:#fff
    style C fill:#e8a838,stroke:#b8842c,color:#fff
    style D fill:#d96a4a,stroke:#a84e35,color:#fff
    style E fill:#7b68ae,stroke:#5a4d82,color:#fff

Quick Navigation

I want to… Start here
Understand the business case Background
See how a run works step by step How It Works
Know what gets checked The 29 Checks
Understand how scores are calculated Scoring
See the impact numbers Results & Impact
Check the project roadmap Roadmap
Look up a term Glossary

This wiki covers the Program Guide (what the agent does and why it matters). A Technical Reference section covering architecture, code modules, and API details for technical contributors is under development and will appear in the sidebar when available.