Topics Directory

Master the foundational and advanced concepts of Machine Learning through structured modules. Prepare for technical interviews with high-density content.

ML Fundamentals

Probability, Bayes, bias-variance, overfitting, cross-validation, and regularization: the statistical core of every ML system

Classical ML

Evaluation metrics, linear and logistic regression, SVMs and kernels, tree ensembles, clustering, and the classical word embeddings that bridged into deep representation learning

Deep Learning Basics

Backprop, activations, initialization, normalization, optimizers, losses, regularization, plus the sequence-modeling arc from n-grams to attention

Pre-Transformer Architectures

CNNs for vision, RNN/LSTM/GRU for sequences, and how attention replaced both

Transformers & Attention

Self-attention, multi-head attention, and the architecture behind modern LLMs

GANs, Diffusion & Flow Matching

The generative-image story from 2014 to 2026: adversarial nets, denoising diffusion, and flow matching

World Models

Generative spatial intelligence: how AI builds and reasons about navigable 3D worlds

Embeddings & Retrieval

Dense vector representations, similarity search, and retrieval systems at scale

Mixture of Experts

Sparse expert architectures that scale model capacity without proportional compute cost

Multimodal Models & VLMs

Vision-language models, multi-modal architectures, and cross-modal understanding

Fine-tuning Strategies

Adapting pre-trained models to specific tasks with full fine-tuning, LoRA, QLoRA, and other PEFT methods

Reinforcement Learning Basics

MDPs, value functions, policy gradients, actor-critic, PPO, and the bridge to RLHF/GRPO for LLMs

RLHF & Alignment

Aligning LLMs with human preferences using RLHF, DPO, and modern preference optimization methods

Retrieval-Augmented Generation

Grounding LLM outputs with external knowledge via retrieval pipelines

Agentic AI

Autonomous AI agents that plan, use tools, and execute multi-step tasks

Vision-Language-Action Models

Foundation models that map camera + instruction directly to robot actions: the bridge from LLMs to physical agency

Test-Time Compute & Reasoning Models

Scaling compute at inference time for reasoning: thinking models like o1, DeepSeek-R1, and Qwen QwQ

Safety, Alignment & Red Teaming

Making LLMs safe, aligned, and robust against adversarial attacks

Distributed Training & Parallelism

How large models are split across many GPUs — data, tensor, pipeline, and ZeRO/FSDP parallelism

Model Quantization

Reducing model size and latency with INT8/INT4 quantization techniques

Model Serving & Inference Optimization

Deploying and optimizing LLM inference at scale with KV caching, batching, and parallelism

Evaluation & Benchmarking

Measuring ML and LLM performance with benchmarks, judges, and production A/B tests

Harness Engineering for AI Agents

Building the closed-loop scaffolding (environment, state, verification, observability) that turns capable models into reliable agents