AI & Machine Learning
Neural networks, supervised learning, NLP, computer vision, and ML engineering.
ML Fundamentals
Supervised vs unsupervised learning, training and validation, overfitting, and evaluation metrics.
Neural Networks
Perceptrons, activation functions, backpropagation, and deep learning architectures.
LLMs & RAG
Large Language Models, Transformers, and Retrieval-Augmented Generation.
Generative AI
Diffusion models, GANs, and multimodal generation (text, image, audio).
Prompt Engineering
Techniques for guiding LLMs: zero-shot, few-shot, and chain-of-thought.
MLOps
Machine Learning Operations, deployment, monitoring, and vector databases.
Models & Fine-Tuning
Hugging Face ecosystem, running models locally, LoRA, and PEFT.
Computer Vision
CNNs, Object Detection (YOLO), Segmentation, and Vision Transformers (ViT).
NLP Fundamentals
Tokenization, Word Embeddings, Named Entity Recognition, and traditional NLP.
Reinforcement Learning
Q-Learning, Policy Gradients, PPO, and autonomous agents.
Local AI Tools
Run LLMs on your own hardware using Ollama, LM Studio, and vLLM.
Multi-Agent Systems
Architectures where multiple specialized AI agents collaborate to solve complex tasks.
RAG Evaluation Metrics
Measuring hallucination, context relevance, and faithfulness in Retrieval-Augmented Generation.
Fine-Tuning SLMs
Adapting Small Language Models (SLMs) for specialized tasks on constrained hardware.
On-Device Inference
Deploying optimized AI models to mobile devices and browsers using WebNN and CoreML.
RLHF & RLAIF
Reinforcement Learning from Human/AI Feedback to align models with intended behaviors.
Multimodal Architectures
Building native multimodal models that simultaneously process text, vision, and audio.
Agentic Tool Use
Equipping LLMs with function calling, code execution, and web browsing capabilities.
Explainable AI (XAI)
Techniques for model interpretability, feature attribution, and mechanistic interpretability.
Federated Learning
Training AI models across decentralized edge devices while preserving data privacy.
Neuromorphic Computing
Brain-inspired hardware and spiking neural networks for ultra-low power AI.
Quantum Machine Learning
Exploring quantum circuits and algorithms to accelerate complex ML tasks.
AI Safety & Alignment
Red-teaming, jailbreak prevention, and ensuring AI systems remain robust and secure.