This article discusses challenges of AI-assisted product changes, particularly context degradation in large language models as conversations progress. It introduces scaffolding, a structured approach to maintain contextual integrity in change management. By emphasizing task decomposition and context engineering, organizations can improve AI performance and enhance governance in engineering workflows.
The post recounts a presentation from the LightX conference, illustrating the importance of Configuration Management (CM) through a personal story involving a school assignment. It highlights how CM, specifically CM2, can prevent miscommunication and inefficiencies in organizations by aligning products, processes, and people, ultimately improving trust and workflow.
In case you missed them, you can find the following How do YOU CM2? posts in the 44th edition of the Future of CM newsletter:
🚫 Stop Calling Product Changes: “Engineering Changes”!
🤫Impact Analysis: The Secret Ingredient No One Wants To Do!
🚀CM2 Baseline: how to make Impact Analysis easy!
This article is part of the How Do YOU CM2? blog series in collaboration with the Institute for Process Excellence (IpX). Although I receive compensation for writing this series, I stand behind its content. I will continue to create and publish high-quality articles that I can fully endorse. Enjoy this new series, and please share your thoughts! In the previous edition of How Do YOU CM2?, we talked about Impact Analysis and why it… Read More »CM2 Baseline: how to make Impact Analysis easy!
Configuration Management relies on baselines, which are approved snapshots of configuration at specific times. They ensure control over changes, provide traceability, and maintain product integrity. Without baselines, organizations face inconsistencies and delays. The CM2 Baseline updates with each dataset release, enhancing visibility and accountability in managing configurations.
On May 15, I presented about the future of configuration management enhanced by AI at the IpX International CM2 Congress. The presentation highlighted a fictitious day with an AI assistant, Chrissy, illustrating its role in tasks like meeting organization, impact analysis, and change planning, while addressing data quality and bias concerns in AI systems.
In a recent IpX True North Podcast, Ray Wozny and Ken Black discussed the importance of Enterprise Change assessment or Impact Analysis for successful change management. They emphasized that effective documentation and ownership are crucial to avoid costly mistakes. Cultural integration of change is essential to ensure sustainable improvements in organizations.
This article discusses enhancing the CM Baseline’s knowledge graph by integrating product and event data, creating a comprehensive ‘product 360.’ It emphasizes the significance of Event Knowledge Graphs to uncover insights overlooked by traditional analysis. By modeling relationships between events and states, it proposes improved predictive capabilities and behavior understanding in product development.
The Model-based Ecosystem committee of the IPX Congress seeks participants for a survey examining the impact of model-based approaches on Configuration Management (CM). The survey, consisting of 11 questions, aims to gather insights on risks and opportunities related to CM processes. Results will be shared on LinkedIn and the Future of CM Blog.
OpenAI’s release of GPT-4 enhances Large Language Models (LLMs) capabilities for impact analysis in configuration management. By leveraging LLMs, knowledge graphs can be utilized to reveal implicit relationships and support decision-making during changes. Upcoming tools, like NeoChat, hint at significant advancements in data exploration for engineers and managers.