PaC: Why IT Management Comes Down to Coding
Posted: Mon Feb 10, 2025 3:24 am
11/13/2023
Two new studies point to the benefits of the Policy-as- Code ( PaC ) model , Computer Weekly reports .
A study conducted by Forrester for Capital One highlights the importance of implementing PaC to optimize software and AI projects. According to Forrester, democratizing machine learning by making ML tools accessible to employees across the organization, including line business units and operations, is critical to accelerating and increasing the contribution of data science to business success.
According to a survey of 180 business data argentina mobile database and decision makers, 67% of respondents believe that one of the main challenges for an AI-focused organization to democratize workflows is that it can increase risk by allowing less data-savvy employees to access ML models and applications. Sixty-three percent believe that a lack of governance or security policies that work effectively with external vendors limits access to ML models and applications, and 64% believe it is difficult to democratize ML while mitigating risk.
Discussing the findings, David Kang, senior vice president and head of Capital One Data Insights, said, “Governance and automation are important aspects of a well-managed data ecosystem.” In his experience, once machine models are in production, automation can help companies ensure continuity of model prediction services. “Automating the monitoring and training of an ML model can ensure that it is healthy when it goes into production and help teams make more accurate decisions about when to retrain a model,” he said. Such automation, Kang said, gives engineers confidence in consistent reproducibility and support.
Two new studies point to the benefits of the Policy-as- Code ( PaC ) model , Computer Weekly reports .
A study conducted by Forrester for Capital One highlights the importance of implementing PaC to optimize software and AI projects. According to Forrester, democratizing machine learning by making ML tools accessible to employees across the organization, including line business units and operations, is critical to accelerating and increasing the contribution of data science to business success.
According to a survey of 180 business data argentina mobile database and decision makers, 67% of respondents believe that one of the main challenges for an AI-focused organization to democratize workflows is that it can increase risk by allowing less data-savvy employees to access ML models and applications. Sixty-three percent believe that a lack of governance or security policies that work effectively with external vendors limits access to ML models and applications, and 64% believe it is difficult to democratize ML while mitigating risk.
Discussing the findings, David Kang, senior vice president and head of Capital One Data Insights, said, “Governance and automation are important aspects of a well-managed data ecosystem.” In his experience, once machine models are in production, automation can help companies ensure continuity of model prediction services. “Automating the monitoring and training of an ML model can ensure that it is healthy when it goes into production and help teams make more accurate decisions about when to retrain a model,” he said. Such automation, Kang said, gives engineers confidence in consistent reproducibility and support.