►What is an AI Engineer?
Free
►Python Environment & Virtualenv
Free
🔒Functions, Modules & Packages
🔒NumPy Arrays & Vectors
🔒Supervised vs Unsupervised
🔒What Are Embeddings?
🔒Neurons, Weights & Biases
🔒Why PyTorch Is the Default
🔒Train / Validation / Test Splits
🔒What Is NLP?
🔒Tokens vs Words vs Characters
🔒Why Transformers Replaced RNNs
🔒Next-Token Prediction
🔒API vs Open-Weight Models
🔒Context Windows Explained
🔒Anatomy of a Good Prompt
🔒Why Structured Output Matters
🔒What Makes a Prompt "Good"?
🔒Why RAG? Limits of LLM Knowledge
🔒From Documents to Embeddings
🔒Document Loading & Parsing
🔒End-to-End RAG Architecture
🔒Hybrid Search (keyword + vector)
🔒Retrieval Metrics (recall, precision)
🔒What Is an AI Agent?
🔒Function / Tool Calling Explained
🔒Orchestration Frameworks (concept)
🔒Fine-tuning vs RAG vs Prompting
🔒Full vs Parameter-Efficient Tuning
🔒Collecting & Curating Data
🔒Building an API with FastAPI
🔒Why Containers? Docker Basics
🔒Token Cost Optimization
🔒Choosing a Capstone Problem
🔒Deploying the Capstone Publicly
🔒GitHub Portfolio Setup
►AI vs ML vs Deep Learning vs GenAI
Free
🔒Notebooks vs Scripts
🔒Working with JSON
🔒Pandas DataFrames
🔒Features & Labels
🔒Turning Text into Vectors
🔒Layers & Activation Functions
🔒Tensors & Operations
🔒Regularization & Dropout
🔒Text Preprocessing
🔒Subword Tokenization (BPE concept)
🔒Self-Attention (intuition)
🔒Pre-training at Scale
🔒Model Sizes & Trade-offs
🔒Input vs Output Tokens
🔒Zero-shot vs Few-shot
🔒Asking for JSON
🔒Building a Test Set
🔒The RAG Pipeline Overview
🔒What Is a Vector Database?
🔒Chunk Size & Overlap
🔒Prompt Assembly with Context
🔒Re-ranking Results
🔒Answer Faithfulness
🔒LLM + Tools + Memory + Loop
🔒Defining Tools & Schemas
🔒Agent Memory & State
🔒What Fine-tuning Can & Cannot Do
🔒LoRA & PEFT (concept)
🔒Formatting Training Examples
🔒Request / Response Design
🔒Building a Container Image
🔒Caching & Batching
🔒Designing the AI Architecture
🔒Building the Demo UI
🔒Showcasing Your Projects
🔒The Modern AI Stack
🔒Managing API Keys & Secrets
🔒Type Hints & Dataclasses
🔒Loading CSV / JSON / Parquet
🔒Classification vs Regression
🔒Semantic Similarity
🔒Forward Propagation
🔒Autograd
🔒Early Stopping
🔒Tokenization Basics
🔒Vocabulary & Token IDs
🔒Multi-Head Attention
🔒What "Parameters" Mean
🔒Latency, Cost & Quality
🔒Managing Long Inputs
🔒Role & System Prompts
🔒JSON Schema & Validation
🔒Manual vs Automated Evaluation
🔒Retrieve, Augment, Generate
🔒Indexing & Similarity Search
🔒Chunking Strategies
🔒Source Citations in Answers
🔒Query Rewriting & Expansion
🔒Groundedness & Citations
🔒Agents vs Workflows
🔒Parsing Tool Calls
🔒Multi-Step Reasoning
🔒Cost & Effort Trade-offs
🔒Instruction Tuning
🔒Cleaning & De-duplication
🔒Streaming to Clients
🔒Secrets in Production
🔒Latency Reduction
🔒RAG / Agent / Fine-tuning Choice
🔒Performance & Cost Check
🔒Writing About Your Work
🔒Where LLMs Fit In
🔒Anatomy of an LLM API Call
🔒Async Basics for API Calls
🔒Cleaning & Missing Values
🔒Training vs Inference
🔒Cosine Similarity & Distance
🔒Loss Functions
🔒Defining a Model (nn.Module)
🔒Hyperparameter Tuning
🔒Bag-of-Words & TF-IDF
🔒Word Embeddings (Word2Vec idea)
🔒Positional Encoding
🔒Emergent Capabilities
🔒Running Open Models Locally (concept)
🔒Truncation & Summarization
🔒Chain-of-Thought Prompting
🔒Function / Tool Schemas
🔒LLM-as-a-Judge (concept)
🔒Grounding & Citations
🔒Metadata Filtering
🔒Embedding the Chunks
🔒Handling "I do not know"
🔒Filtered Retrieval
🔒Building a RAG Test Set
🔒The ReAct Pattern (reason + act)
🔒Executing & Returning Results
🔒Multi-Agent Patterns
🔒Use Cases for Fine-tuning
🔒Quantization Basics
🔒Train / Validation Split
🔒Building UIs (Streamlit / Gradio)
🔒Deploying to the Cloud (concept)
🔒Input / Output Guardrails
🔒Building the Core Pipeline
🔒Recording a Demo Video
🔒Interview Prep for AI Roles
🔒Capabilities & Limits of GenAI
🔒Requests, Responses & Tokens
🔒Environment Variables
🔒Filtering & Aggregation
🔒Overfitting & Generalization
🔒Embedding Models (concept)
🔒Gradient Descent (intuition)
🔒The Training Loop
🔒Reading Loss Curves
🔒Sequence Modeling Problems
🔒Contextual Embeddings
🔒Encoder vs Decoder
🔒Strengths & Failure Modes
🔒Multimodal Models
🔒System / User / Assistant Roles
🔒Instruction Clarity
🔒Parsing & Error Recovery
🔒Versioning Prompts
🔒RAG vs Fine-tuning
🔒Hosted vs Self-Hosted
🔒Top-K Retrieval
🔒Conversation + Retrieval
🔒Multi-Query Retrieval
🔒Detecting Hallucinations
🔒Planning & Decomposition
🔒Multi-Tool Selection
🔒Human-in-the-Loop
🔒Data Requirements
🔒Hardware Considerations
🔒Evaluating a Fine-tuned Model
🔒Authentication Basics
🔒Scaling Basics
🔒Prompt-Injection Awareness
🔒Iterating with Evaluation
🔒Writing the Case Study
🔒Building Your Personal Brand
🔒Common AI Application Patterns
🔒Temperature & Sampling
🔒Logging
🔒Basic Visualization
🔒Accuracy, Precision & Recall
🔒Visualizing Embeddings
🔒Back-Propagation (intuition)
🔒Optimizers & Schedulers
🔒Transfer Learning (intro)
🔒Classic vs Modern NLP
🔒Token Limits & Context
🔒Pre-training vs Fine-tuning
🔒Hallucinations Explained
🔒How to Choose a Model
🔒Conversation Memory
🔒Delimiters & Formatting
🔒Constrained Generation (concept)
🔒A/B Testing Prompts
🔒RAG Architecture Patterns
🔒Choosing a Vector Store
🔒Relevance & Recall
🔒Caching
🔒Hierarchical / Parent-Child Chunks
🔒Continuous Evaluation
🔒When Not to Use Agents
🔒Error Handling in Tool Use
🔒Observability & Tracing
🔒Risks & Pitfalls
🔒Hosted Fine-tuning Services
🔒Avoiding Catastrophic Forgetting
🔒Async & Concurrency
🔒CI/CD Overview
🔒Logging & Observability
🔒Documentation & README
🔒The CareerByteCode Showcase
🔒Responsible & Ethical AI
🔒Streaming Responses
🔒Reading & Writing Files
🔒Train / Test Concepts
🔒The Bias-Variance Idea
🔒Search, Clustering & Recommendations
🔒Epochs, Batches & Learning Rate
🔒GPU vs CPU
🔒Model Evaluation in Practice
🔒Classification, NER, Summarization
🔒Counting Tokens for Cost
🔒From Transformer to GPT-style Models
🔒Determinism & Randomness
🔒Examples: GPT, Claude, Llama, Mistral
🔒Token Budgeting & Cost Control
🔒Common Prompting Pitfalls
🔒Output Reliability Techniques
🔒Reducing Hallucinations
🔒Real-World RAG Use Cases
🔒Examples: Pinecone, Chroma, pgvector, FAISS
🔒Handling Tables, PDFs & Mixed Content
🔒Putting a UI on RAG
🔒Reducing Irrelevant Context
🔒Common RAG Failure Modes
🔒Agent Use Cases
🔒Safety & Guardrails for Actions
🔒Examples: LangChain, LlamaIndex
🔒A Decision Framework
🔒Open-Model Fine-tuning (concept)
🔒Iterating Safely
🔒Connecting Front-end to AI Back-end
🔒Free Hosting Options for Demos
🔒Monitoring Quality & Privacy
🔒Adding Guardrails
🔒The AI Product Lifecycle
🔒Rate Limits & Error Handling
🔒Calling REST APIs
🔒Working with Text Data
🔒Why GenAI Still Needs ML Thinking
🔒Limitations of Embeddings
🔒Why Deep Networks Work
🔒Saving & Loading Models
🔒Why Prompts Matter
🔒Reusable Prompt Templates
🔒Roles in an AI Team
🔒Cost Awareness
🔒Clean Code for AI Apps
🔒Data Quality for AI
🔒Datasets & DataLoaders
🔒Industry Use Cases
🔒Examples: OpenAI, Claude, Open Endpoints