AI in Context
Making AI simple, practical, and useful.
AI in Context is an ongoing project. This section will grow continuously as I continue my journey: exploring, experimenting, building, and sharing.
Mission Statement
The project "AI in Context" explores how artificial intelligence works, where it fits, and how different computational approaches can be combined to build useful systems.
The emphasis is on understanding the fundamentals, building practical systems, experimenting with ideas, and learning what actually works.
Major Areas Covered
The AI in Context project is organised around five major areas:
- Back to Basics
- Learning & Decisions
- AI Learning Models
- AI Applications
- AI Systems & Engineering
These areas provide a broad framework for the project. They are not intended as a fixed curriculum; material will be added and developed organically as the project progresses.
1. Back to Basics
The foundations beneath AI and intelligent computing.
- Mathematics
- Programming
- Algorithms
- Optimization
- Computing
- Other fundamentals
2. Learning & Decision Methods
Different ways computational systems can learn, adapt, search, and make decisions.
- Machine Learning
- Supervised and Unsupervised Learning
- Reinforcement Learning
- Evolutionary Computation
- Search & Planning
- Game Theory
- Other Decision Methods
3. Representation & AI Models
How information, knowledge, and patterns are represented and processed by intelligent systems.
- Neural Networks
- Deep Learning
- Transformers
- Embeddings & Representations
- Knowledge Representation
- Symbolic Methods
- Vision & Language Models
- Multimodal Models
- Other AI Models
4. Modern AI & Applications
Modern AI capabilities and the problems they can help solve.
- Large Language Models (LLMs)
- Generative AI
- Machine Vision
- Natural Language Processing
- Speech & Audio
- Multimodal AI
- AI Reasoning
- AI for Education
- AI for Science & Research
- AI-Assisted Programming
- Other Applications
5. AI Systems & Engineering
Turning AI models and computational methods into useful, reliable systems.
- AI Agents
- Backend & APIs
- Data & Databases
- Retrieval & RAG
- Memory & Context
- Tools & Orchestration
- Workflows & Automation
- Model Serving & Inference
- Distributed Computing
- Evaluation & Reliability
- User Interfaces
- Deployment & Operations
How the Pieces Fit Together
AI in Context follows a practical progression:
Foundations → Learning → Representation → Modern AI & Applications → Systems
The boundaries are deliberately not rigid.
A useful AI system may combine several approaches: for example, neural models with search, optimization, symbolic reasoning, evolutionary methods, or game-theoretic decision making.
The aim is not to follow one particular AI paradigm, but to understand the different tools available and learn how to combine them effectively.
Principles
- Understand before using.
- Build before over-theorizing.
- Learn the fundamentals behind the abstractions.
- Use today's technology to solve today's problems.
- Experiment and measure.
- Prefer useful systems over impressive demonstrations.
- Stay adaptable as technology changes.