The list of terminologies is not exhaustive. It is being updated regularly.
Search AI Terms
Flashcard
Classification criteria used to automate categorizing the terminologies into Beginner, Intermediate and Advanced Levels.
PURPOSE OF CLASSIFICATION
To assign each AI term a Knowledge Level based on the depth of understanding, technical complexity, and real-world applicability required for a Product Manager to effectively use or discuss it in product-related contexts.
KNOWLEDGE LEVELS
| Level | Name | Description |
|---|---|---|
| 1 | Beginner | High-level awareness. PM should understand what the term is, why it matters, and its basic use case. No technical background required. |
| 2 | Intermediate | Functional understanding. PM should know when to apply the concept, understand its value, discuss trade-offs, and interpret basic results. Some collaboration with technical teams is expected. |
| 3 | Advanced | Deeper technical fluency. PM should grasp how the concept works, influence system design decisions, and interact closely with data scientists or engineers. Useful for AI-first products or technical PMs. |
CLASSIFICATION PARAMETERS
These are mutually exclusive and collectively exhaustive criteria to determine the appropriate level:
| Parameter | Beginner | Intermediate | Advanced |
|---|---|---|---|
| Conceptual Simplicity | Easily understood in one sentence or analogy | Requires layered understanding | Abstract, multi-layered, or math-heavy |
| Real-World Frequency | Commonly used in business/product settings | Less common but important in specific domains | Rare or used in cutting-edge or research-heavy areas |
| Technical Depth Needed | No coding/math knowledge needed | Familiarity with ML workflows, metrics, or tooling | Understanding of algorithms, architectures, or optimization techniques |
| Collaboration Role | PM can discuss outcomes and use cases | PM can define requirements or ask for specific outputs | PM influences model selection, architecture, or evaluation strategy |
| Decision-Making Involvement | Helps frame product/business value | Helps define how to solve the problem | Helps define how a model learns or performs |
| Data Science Dependency | Independent or surface-level dependency | Requires translating between business & data science | Closely tied to data scientist’s technical work |
| Documentation Availability | Widely explained in non-technical language | Requires some domain-specific reading | Mostly explained through research papers or advanced blogs |
| Tooling & Platforms | Available in standard PM tools (AutoML, LLM APIs) | Part of workflows like Jupyter notebooks or dashboards | Requires understanding frameworks (e.g., PyTorch, TensorFlow) |