All Episodes

Context Engineering

Manus AI

Kimi K2

Mixture-of-Recursions (MoR)

MeanFlow

Mamba

LLM Alignment

Why We Think

Deep Research

vLLM

Qwen3: Thinking Deeper, Acting Faster

RAGEN: train and evaluate LLM agents using multi-turn RL

DeepSeek-Prover-V2

DeepSeek-Prover

Model Context Protocol (MCP)

LLM Post-Training: Reasoning

Agent AI Overview

FlashAttention-3

FlashAttention-2

FlashAttention

PPO (Proximal Policy Optimization)

"Deep Dive into LLMs like ChatGPT" - Andrej Karpathy's Tech Talk Learning

"Intro to Large Language Models" - Andrej Karpathy's Tech Talk Learning

DeepSeek-V2

Matrix Calculus in Deep Learning

S1: Simple Test-time Scaling

RLHF (Reinforcement Learning from Human Feedback)

GRPO (Group Relative Policy Optimization)

Model/Knowledge Distillation

Qwen-2.5

Qwen-2

Qwen-1

OpenAI-o1

GPT-4o

Kimi k1.5

DeepSeek-R1

Claude-3

GPT-4

LLM Training

MiniMax-01

DeepSeek v3

Tree-of-Thoughts

LLM Reasoning

LangChain

LlamaIndex

Chain of Thought (CoT)

Retrieval-Augmented Generation (RAG)

Fine-Tuning

Scaling Laws

LLaMA-3

LLaMA-2

LLaMA-1

Survey of Large Language Models

Mixture of Experts (MoE)

Multi-Task Learning

Gradient Descent Optimization Algorithms

GPT-1 (Generative Pre-trained Transformer)

Linear Transformers

BERT

Sora

Word2Vec

Stable Diffusion

Retrieval Transformer

GPT-2

GPT-3

Transformer

Prompt Engineering

Agentic AI