KNOWLEDGE BASE
Glossaries
A shared vocabulary across strategy, technology, and transformation
Knowledge Base
Glossary of Key Terms
A shared vocabulary for strategy, technology, and transformation.
Clear language is essential for effective collaboration across business, technology, and change teams. This glossary defines the core terms we use in our consulting, training, and advisory work — from AI and cybersecurity to agile delivery and systems thinking.
AI & Machine Learning
Core concepts in artificial intelligence, large language models, and enterprise AI deployment that shape how organizations adopt and scale intelligent systems.
- Large Language Model (LLM)
- A deep learning model trained on massive text corpora that can generate, summarize, translate, and reason over natural language. LLMs such as GPT-4, Claude, and Llama form the foundation of modern generative AI applications in the enterprise.
- Retrieval-Augmented Generation (RAG)
- An architecture pattern that combines a language model with an external knowledge retrieval step. RAG grounds model outputs in verified documents, reducing hallucination and enabling domain-specific answers without full model retraining.
- Fine-Tuning
- The process of further training a pre-trained model on a domain-specific dataset to improve its performance on specialized tasks. Fine-tuning balances customization with the general capabilities inherited from the base model.
- On-Premises AI
- AI systems deployed within an organization's own data centers or private infrastructure rather than in public cloud environments. On-premises deployment addresses data sovereignty, latency, and regulatory compliance requirements that public cloud cannot satisfy.
- Agentic AI
- AI systems capable of autonomous goal-directed behavior — planning multi-step tasks, using tools, and adapting strategy based on feedback. Agentic AI moves beyond single-turn question answering toward persistent, delegated workflows within enterprise processes.
- Prompt Engineering
- The practice of crafting inputs to language models to elicit accurate, relevant, and well-structured outputs. Effective prompt engineering involves techniques such as few-shot examples, chain-of-thought reasoning, and system-level instructions.
- Foundation Model
- A large-scale AI model pre-trained on broad data that serves as the starting point for many downstream tasks. Foundation models can be adapted through fine-tuning, RAG, or prompt engineering to serve diverse enterprise use cases from a single base.
- MLOps
- A set of practices for deploying, monitoring, and maintaining machine learning models in production. MLOps covers the full lifecycle from experiment tracking and model versioning to automated retraining, drift detection, and observability.
Agile & Delivery
Frameworks, roles, and practices for iterative delivery and organizational agility that help teams and leaders deliver value faster with higher quality.
- Scrum
- A lightweight framework for developing and sustaining complex products through iterative cycles called Sprints. Scrum defines three roles (Product Owner, Scrum Master, Developers), five events, and three artifacts that create transparency and continuous improvement.
- Kanban
- A method for managing knowledge work with an emphasis on visualizing the flow, limiting work in progress, and optimizing throughput. Kanban helps teams identify bottlenecks and improve delivery predictability without prescribing fixed iteration cycles.
- Sprint
- A time-boxed iteration, typically one to four weeks, during which a Scrum team produces a usable product increment. Each Sprint begins with planning and ends with a review and retrospective to inspect and adapt both the product and the process.
- Product Owner
- The person accountable for maximizing the value of the product by managing and ordering the Product Backlog. The Product Owner represents stakeholders and customers, making trade-off decisions that balance business value, risk, and technical feasibility.
- Scrum Master
- A servant-leader who helps the Scrum Team and the wider organization understand and apply Scrum effectively. The Scrum Master facilitates events, removes impediments, coaches self-management, and helps the organization adopt agile practices.
- SAFe (Scaled Agile Framework)
- An enterprise framework for scaling agile and lean practices across large organizations. SAFe provides structured guidance for portfolio, program, and team-level alignment through Agile Release Trains, PI Planning, and lean governance mechanisms.
- Agile Coaching
- Professional facilitation and mentoring that helps individuals, teams, and organizations adopt and improve agile ways of working. Agile coaches work across mindset, practices, and organizational design to create lasting systemic change.
- Backlog Refinement
- An ongoing activity in which the Product Owner and the development team review, clarify, estimate, and prioritize items in the Product Backlog. Effective refinement ensures the team always has a well-understood, ordered set of work ready for upcoming Sprints.
Systems Thinking
Principles and models for understanding complex, interconnected systems — the intellectual foundation for systemic consulting and organizational design.
- System Dynamics
- A methodology for studying the behavior of complex systems over time using stocks, flows, feedback loops, and time delays. System dynamics helps leaders understand why interventions often produce counter-intuitive results and how to design more effective policies.
- Feedback Loop
- A circular causal chain in which a system's output feeds back as an input, either reinforcing the original trend (positive feedback) or counteracting it (negative feedback). Understanding feedback loops is essential for diagnosing systemic problems and designing stable interventions.
- Emergence
- The phenomenon by which macro-level patterns, behaviors, or properties arise from the interactions of simpler components that do not individually exhibit those properties. Emergence explains why organizations behave in ways that cannot be predicted from the behavior of individual teams alone.
- Complexity
- A property of systems characterized by many interdependent agents, nonlinear relationships, and unpredictable outcomes. In complex environments, traditional command-and-control management fails; leaders must instead probe, sense, and respond through safe-to-fail experiments.
- Viable System Model (VSM)
- A model of the organizational structure of any autonomous system capable of producing itself, developed by Stafford Beer. The VSM identifies five necessary and sufficient functions that any viable system must perform to maintain its identity and adapt to its environment.
- Soft Systems Methodology (SSM)
- An approach to tackling ill-defined, real-world problems by building conceptual models of human activity systems and comparing them with perceived reality. SSM is particularly useful in situations where stakeholders hold conflicting worldviews about what the problem actually is.
Cybersecurity
Approaches and concepts for protecting enterprise systems, data, and infrastructure in an increasingly threat-rich and regulation-heavy environment.
- Zero Trust
- A security model based on the principle that no user, device, or network segment should be trusted by default — every access request must be explicitly verified, authorized, and continuously validated regardless of origin. Zero Trust is especially critical for organizations with distributed workforces and hybrid cloud environments.
- Threat Modeling
- A structured process for identifying, quantifying, and prioritizing potential security threats to a system or application. Effective threat modeling happens early in the design phase and considers attacker profiles, attack surfaces, data flows, and trust boundaries.
- Penetration Testing
- Authorized simulated attacks against a system to evaluate its security posture and discover exploitable vulnerabilities before real adversaries do. Penetration tests complement automated scanning by applying human creativity and contextual judgment to realistic attack scenarios.
- Security by Design
- An approach where security requirements and controls are integrated into every phase of the software development lifecycle rather than bolted on after the fact. Security by design reduces remediation costs, minimizes attack surface, and accelerates compliance readiness.
- Incident Response
- The organized approach to detecting, containing, eradicating, and recovering from cybersecurity events. A mature incident response capability reduces mean time to detection and recovery, limits business impact, and satisfies regulatory notification obligations.
- Data Residency
- The requirement that data be stored, processed, and managed within specific geographic boundaries as dictated by regulations, contractual obligations, or organizational policy. Data residency is a key driver for on-premises AI deployment in heavily regulated sectors.
Let’s apply these concepts to your organization
Want to explore how these concepts come alive in your transformation journey? Reach out to our advisory team.
Get in Touch