Foundations
What AI Actually Is
- AI vs machine learning vs deep learning
- Narrow AI vs general AI
- Where modern AI fits in real products
Coming soon
Learning Path
AI Roadmap
A progressive path to understand modern AI, from core concepts to production systems built with LLMs, RAG, agents, and serving stacks.
Foundations
Math and Statistics You Need
- Vectors and matrices intuition
- Probability basics
- Train, validation, and test mindset
Coming soon
Data
Data Quality and Preparation
- Structured vs unstructured data
- Cleaning and labeling basics
- Bias and leakage risks
Coming soon
Models
Supervised Learning Basics
- Classification vs regression
- Features and targets
- Overfitting and underfitting
Coming soon
Models
Unsupervised Learning and Embeddings
- Clustering basics
- Similarity search intuition
- Embeddings in modern AI systems
Coming soon
LLMs
Transformer and Token Mental Model
- Tokens and context window
- Attention intuition
- Inference vs training
Coming soon
Prompting
Prompt Engineering Fundamentals
- Clear instructions
- Role, context, and examples
- Output constraints and delimiters
Coming soon
LLMs
Model Selection and Tradeoffs
- Quality vs latency vs cost
- Hosted vs local models
- Open weights vs closed APIs
Coming soon
RAG
Retrieval-Augmented Generation
- Chunking and indexing
- Vector search basics
- Grounding answers in source data
Coming soon
Evaluation
Evals and Failure Analysis
- Golden datasets
- Hallucination patterns
- Measure task success
Coming soon
Safety
Safety, Privacy, and Guardrails
- PII and sensitive data handling
- Prompt injection basics
- Moderation and policy checks
Coming soon
Agents
Tools, Functions, and Agents
- Function calling
- Tool orchestration
- When agents are actually useful
Coming soon
Serving
Inference APIs and Serving
- Request and response design
- Batching and streaming
- Rate limits and retries
Coming soon
Serving
Local AI Stack
- Ollama and local runtimes
- CPU and GPU constraints
- Private deployment patterns
Coming soon
Product
UX for AI Features
- Human-in-the-loop design
- Fallback and uncertainty handling
- Prompt and response UX
Coming soon
MLOps
Versioning and Experiment Tracking
- Prompt and model versioning
- Dataset reproducibility
- Experiment logs and comparisons
Coming soon
Production
Monitoring and Cost Control
- Latency, token, and error metrics
- Drift and quality regressions
- Cache and spend management
Coming soon
Production
Production AI Systems
- RAG plus tools plus auth
- Deployment patterns
- Iterate safely over time
Coming soon