amitness/learning
amitness/learning là dự án tích hợp AI trên GitHub với 7.0k sao. A log of things I'm learning
Tóm tắt dựng từ metadata GitHub của chính dự án — chưa có bài review TopGit. Trang sẽ tự động cập nhật khi bài review đầy đủ được xuất bản.
TopGit viết bài đầy đủ cho repo có nhiều sao nhất và được yêu cầu nhiều nhất. Trang này là snapshot trong thời gian chờ — xem README gốc ở tab READ ME.
Snapshot
Cộng tác viên hàng đầu
Xem cộng tác viên hàng đầu
learning
A running log of things I'm learning to build strong fundamentals while also expanding my knowledge of adjacent technologies everyday.
Updated: Once a month | Current Focus: Generative AI
Generalized Skills
Domain-agnostic, stack-independent skills
System Design
| Format | Resource | Length | Progress |
|---|---|---|---|
| Book | Designing Machine Learning Systems | 386 pgs | ✅ |
| Designing Data-Intensive Applications | 616 pgs | ⬜ | |
| Udacity | A/B Testing | 5/5 | ⬜ |
| Authentication & Authorization: OAuth | 4/4 | ✅ | |
| Client-Server Communication | 5/5 | ⬜ | |
| Designing RESTful APIs | 5/5 | ✅ | |
| HTTP & Web Servers | 3/3 | ✅ | |
| Networking for Web Developers | 5/5 | ✅ | |
| Udemy | AWS Certified Developer - Associate 2018 | ✅ | |
| Datacamp | A/B Testing in Python | 4h | ✅ |
| Customer Analytics & A/B Testing in Python | 4h | ✅ | |
| Machine Learning Monitoring Concepts | 2h | ✅ | |
| MLOps Concepts | 2h | ✅ | |
| Neetcode | System Design for Beginners | ✅ | |
| System Design Interview | ✅ |
Maths
| Format | Resource | Length | Progress |
|---|---|---|---|
| Datacamp | Foundations of Probability in Python | 5h | ✅ |
| Introduction to Statistics | 4h | ✅ | |
| Introduction to Statistics in Python | 4h | ✅ | |
| Hypothesis Testing in Python | 4h | ✅ | |
| Statistical Thinking in Python (Part 1) | 3h | ✅ | |
| Statistical Thinking in Python (Part 2) | 4h | ✅ | |
| Experimental Design in Python | 4h | ✅ | |
| edX | Essential Statistics for Data Analysis using Excel | ✅ | |
| MIT | MIT 18.06 Linear Algebra, Spring 2005 | 36/36 | ✅ |
| Udacity | Intro to Inferential Statistics | 23/23 | ✅ |
| Eigenvectors and Eigenvalues | 1/1 | ✅ | |
| Linear Algebra Refresher | 8/8 | ⬜ | |
| Youtube | Essence of linear algebra | 16/16 | ⬜ |
Data Structures and Algorithms
| Format | Resource | Length | Progress |
|---|---|---|---|
| Book | Grokking Algorithms | 254 pgs | ✅ |
| A Common-Sense Guide to Data Structures and Algorithms | 489/489 pgs | ✅ | |
| Udacity | Intro to Data Structures and Algorithms | ✅ | |
| Neetcode | Algorithms and Data Structures for Beginners | ✅ | |
| Advanced Algorithms | 1/7 | ⏳ | |
| Youtube | Sliding Window Technique - Algorithmic Mental Models | 36m | ✅ |
Data Modeling
| Format | Resource | Length | Progress |
|---|---|---|---|
| Udacity | Intro to relational database | 4/4 | ✅ |
| SQL for Data Analysis | 10/10 | ⬜ | |
| Database Systems Concepts & Design | ⬜ | ||
| Datacamp | Database Design | 4h | ✅ |
| Introduction to Databases in Python | 4h | ⬜ | |
| Intro to SQL for Data Science | 2h | ✅ | |
| Intermediate SQL | 4h | ✅ | |
| Joining Data in SQL | 4h | ✅ | |
| Data Manipulation in SQL | 4h | ⬜ | |
| Exploratory Data Analysis in SQL | 4h | ⬜ | |
| Analyzing Business Data in SQL | 4h | ⬜ | |
| Reporting in SQL | 4h | ⬜ | |
| NoSQL Concepts | 2h | ✅ | |
| Introduction to MongoDB in Python | 3h | ⬜ |
Design Sense
| Format | Resource | Length | Progress |
|---|---|---|---|
| Book | Refactoring UI | 213 pgs | ⬜ |
| Pluralsight | UX Fundamentals | ✅ | |
| Course | How to Visualize Value | ✅ | |
| Series: K-12 Figma Design Basics | ✅ | ||
| Youtube | How to Make Your Website Not Ugly: Basic UX for Programmers | 48m | ✅ |
| Article | Create an illustration in Figma design | ✅ |
Linux & Command Line
| Format | Resource | Length | Progress |
|---|---|---|---|
| Udacity | Linux Command Line Basics | 30/30 | ✅ |
| Shell Workshop | 1/1 | ✅ | |
| Configuring Linux Web Servers | 3/3 | ✅ | |
| Datacamp | Introduction to Shell for Data Science | 4h | ✅ |
| Introduction to Bash Scripting | 4h | ✅ | |
| Data Processing in Shell | 4h | ✅ | |
| MIT | The Missing Semester | 11/11 | ✅ |
| Article | A guide to manage your environment variables in a better way using direnv | ✅ |
Version Control
| Format | Resource | Length | Progress |
|---|---|---|---|
| Udacity | Version Control with Git | 6/6 | ✅ |
| GitHub & Collaboration | ✅ | ||
| How to Use Git and GitHub | 4/4 | ✅ | |
| Datacamp | Introduction to Git for Data Science | 4h | ✅ |
| Advanced Git | 3h | ✅ | |
| Youtube | How to Use Git Worktree | Checkout Multiple Git Branches at Once | 11m | ✅ |
Testing & Profiling
| Format | Resource | Length | Progress |
|---|---|---|---|
| Datacamp | Unit Testing for Data Science in Python | 4h | ✅ |
| Youtube | Debug Python inside Docker using debugpy and VSCode | 8m | ✅ |
| Improving Your Tests With the Python Mock Object Library: Introducing Mock & Replacing Code | 16m | ✅ | |
| Hot-swapping Symbols During Tests: Demystifying unittest.mock[.patch] - Bryce Beagle | 30m | ✅ | |
| Article | Introduction to Memory Profiling in Python | ✅ | |
| Profiling Python code with memory_profiler | ✅ | ||
| How to Use "memory_profiler" to Profile Memory Usage by Python Code? | ✅ | ||
| Concurrency For Starlette Apps (e.g FastAPI / FastHTML) | ✅ |
Technical Writing
| Format | Resource | Length | Progress |
|---|---|---|---|
| Course | Google: Technical Writing One | ✅ | |
| Google: Technical Writing Two | ⬜ | ||
| Compound Content | ✅ | ||
| Build Once, Sell Twice | ✅ | ||
| Book | Writing for Software Developers | ✅ |
Career
| Format | Resource | Length | Progress |
|---|---|---|---|
| Book | The Tech Resume Inside Out | 280 pgs | ✅ |
| The Coding Career Handbook | ✅ | ||
| The Software Engineer's Guidebook | 499 pgs | ✅ |
Programming
| Format | Resource | Length | Progress |
|---|---|---|---|
| Book | Python 201 | 352 pgs | ⬜ |
| Writing Idiomatic Python 3 | 332 pgs | ✅ | |
| Datacamp | Writing Efficient Python Code | 4h | ✅ |
| Writing Functions in Python | 4h | ✅ | |
| Object-Oriented Programming in Python | 4h | ✅ | |
| Intermediate Object-Oriented Programming in Python | 4h | ✅ | |
| Importing Data in Python (Part 1) | 3h | ✅ | |
| Importing Data in Python (Part 2) | 2h | ✅ | |
| Intermediate Python for Data Science | 4h | ✅ | |
| Python Data Science Toolbox (Part 1) | 3h | ✅ | |
| Python Data Science Toolbox (Part 2) | 4h | ✅ | |
| Developing Python Packages | 4h | ✅ | |
| Conda Essentials | 2h | ✅ | |
| Working with Dates and Times in Python | 4h | ✅ | |
| Youtube | Tutorial: Sebastian Witowski - Modern Python Developer's Toolkit | 2h24m | ✅ |
| Article | Python's many command-line utilities | ⬜ | |
| A Programmer’s Introduction to Unicode | ✅ | ||
| Exposing string types to maximize user happiness | ✅ |
Development Environment
| Format | Resource | Length | Progress |
|---|---|---|---|
| DeepLearning.AI | Build Apps with Windsurf’s AI Coding Agents | 1h30m | ✅ |
| Pluralsight | Using The Chrome Developer Tools | ✅ | |
| Youtube | Prompt Driven Development Series | 9/9 | ✅ |
| How Anthropic Built Claude Code (w/ Boris Cherny) | 50m | ✅ | |
| Docs | VSCode Docs: Python Interactive window | ✅ | |
| Article | A Guide to Claude Code 2.0 and getting better at using coding agents | ✅ | |
| Understanding is the new bottleneck | ✅ |
Specialized Skills
Generative AI
Natural Language Processing
| Format | Resource | Length | Progress |
|---|---|---|---|
| Book | Book: Natural Language Processing with Transformers | 400 pgs | ✅ |
| Stanford | CS224U: Natural Language Understanding | Spring 2019 | 15/15 | ✅ |
| CS224N: NLP with Deep Learning | Winter 2019 | 22/22 | ✅ | |
| CMU | Low-resource NLP Bootcamp 2020 | 8/8 | ✅ |
| Multilingual NLP 2020 | ✅ | ||
| Datacamp | Feature Engineering for NLP in Python | 4h | ✅ |
| Natural Language Processing Fundamentals in Python | 4h | ✅ | |
| Regular Expressions in Python | 4h | ✅ | |
| RNN for Language Modeling | 4h | ✅ | |
| Natural Language Generation in Python | 4h | ✅ | |
| Building Chatbots in Python | 4h | ✅ | |
| Sentiment Analysis in Python | 4h | ✅ | |
| Machine Translation in Python | 4h | ✅ | |
| Article | The Unreasonable Effectiveness of Collocations | ⬜ | |
| FuzzyWuzzy: Fuzzy String Matching in Python | ✅ | ||
| Transformers: Origins | ✅ | ||
| Understanding the Difference Between Embedding Layers and Linear Layers | ✅ |
LLM Theory
| Format | Resource | Length | Progress |
|---|---|---|---|
| Book | Hands-On Large Language Models: Language Understanding and Generation | 424 pgs | ✅ |
| Large Language Models: A Deep Dive: Bridging Theory and Practice | 600 pgs | ⬜ | |
| Build a Large Language Model (From Scratch) | 472 pgs | ✅ | |
| The Hundred-Page Language Models Book | 100 pgs | ✅ | |
| Super Study Guide: Transformers & Large Language Models | 350 pgs | ✅ | |
| The Smol Training Playbook: The Secrets to Building World-Class LLMs | ⬜ | ||
| Courses | Stanford CS336: Language Modeling from Scratch | 18/18 | ✅ |
| DeepLearning.AI | Pretraining LLMs | 1h29m | ✅ |
| Reinforcement Learning from Human Feedback | 1h32m | ✅ | |
| How Transformer LLMs Work | 1h44m | ✅ | |
| Karpathy | Intro to Large Language Models | 1h | ✅ |
| Let's build the GPT Tokenizer | 2hr13m | ✅ | |
| Let's reproduce GPT-2 (124M) | 4hr1m | ✅ | |
| Deep Dive into LLMs like ChatGPT | 3h31m | ✅ | |
| Youtube | 5 Years of GPTs with Finbarr Timbers | 55m | ✅ |
| A Hackers' Guide to Language Models | 1h30m | ✅ | |
| Stanford CS229 I Machine Learning I Building Large Language Models (LLMs) | 1h44m | ✅ | |
| LLaMA explained: KV-Cache, Rotary Positional Embedding, RMS Norm, Grouped Query Attention, SwiGLU | 1h10m | ✅ | |
| CMU Advanced NLP Fall 2024 (14): Ensembling and Mixture of Experts | 1h5m | ✅ | |
| A little guide to building Large Language Models in 2024 | 1h15m | ✅ | |
| How I use LLMs | 2h7m | ✅ | |
| Simple Diffusion Language Models | 15m | ✅ | |
| Zed Inferred: Diffusion Language Models | 1h2m | ✅ | |
| Some open problems I am currently thinking about | 30m | ✅ | |
| Rotary Positional Embeddings: Combining Absolute and Relative | 11m | ✅ | |
| How Rotary Position Embedding Supercharges Modern LLMs [RoPE] | 13m | ✅ | |
| Large Language Models in Five Formulas | 58m | ✅ | |
| What I Learned From Implementing LLM Architectures From Scratch (And How to Get Started) | 52m | ✅ | |
| LLM Building Blocks & Transformer Alternatives | 27m | ✅ | |
| Everything I Learned Training Frontier Small Models | 20m | ✅ | |
| The Secrets to Training World-Class LLMs | 1h33m | ✅ | |
| Article | You could have designed state of the art Positional Encoding | ✅ | |
| From Digits to Decisions: How Tokenization Impacts Arithmetic in LLMs | ✅ | ||
| SolidGoldMagikarp (plus, prompt generation) | ✅ | ||
| Sampling for Text Generation | ✅ | ||
| First Token Cutoff LLM sampling | ✅ | ||
| The Big LLM Architecture Comparison | ✅ | ||
| From GPT-2 to gpt-oss: Analyzing the Architectural Advances | ✅ | ||
| A Visual Guide to Mamba and State Space Models | ⬜ | ||
| Patterns and Messages - Part 1 - The Missing Subscript | ✅ | ||
| How text diffusion works | ✅ | ||
| The Illustrated Evo 2 | ⬜ | ||
| Interpreting the Prediction of BERT Model for Text Classification | ✅ | ||
| 2025: The year in LLMs | ✅ | ||
| Things we learned about LLMs in 2024 | ✅ | ||
| Synthetic pretraining | ✅ | ||
| The Novice's LLM Training Guide | ✅ | ||
| The Potential of RLMs | ✅ | ||
| The Curious Case of the bos_token | ✅ | ||
| Transformer Math 101 | ✅ |
Post-training (RLHF / RLVR)
| Format | Resource | Length | Progress |
|---|---|---|---|
| Book | A Little Bit of Reinforcement Learning from Human Feedback | 280 pgs | ✅ |
| DeepLearning.AI | Post-training of LLMs | 1h26m | ✅ |
| Youtube | How DeepSeek Changes the LLM Story | 49m | ✅ |
| Speculations on Test-Time Scaling (o1) | 47m | ✅ | |
| DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning | 1h19m | ✅ | |
| MIT EI seminar, Hyung Won Chung from OpenAI. "Don't teach. Incentivize." | 35m | ✅ | |
| Group Relative Policy Optimization (GRPO) - Formula and Code | 24m | ✅ | |
| How to approach post-training for AI applications | 22m | ✅ | |
| Building Cursor Composer with Sasha Rush | 19m | ✅ | |
| The RLHF Book Course | 5/5 | ⬜ | |
| Full Workshop: Reinforcement Learning, Kernels, Reasoning, Quantization & Agents — Daniel Han | 2h42m | ✅ | |
| Special Topics in Kernels, RL, Reward Hacking in Agents — Daniel Han, Unsloth | 2h20m | ✅ | |
| Everything Is a Rollout — Alex Shaw + Ryan Marten, Terminal-Bench, Harbor | 21m | ✅ | |
| Article | Scaling test-time compute - a Hugging Face Space by HuggingFaceH4 | ✅ | |
| DeepSeek R1's recipe to replicate o1 and the future of reasoning LMs | ✅ | ||
| The Illustrated DeepSeek-R1 | ✅ | ||
| A Visual Guide to Reasoning LLMs | ✅ | ||
| GRPO in DeepSeek-R1 | ✅ | ||
| PPO & GRPO for LLM Alignment | ✅ | ||
| Introduction to LLM Post-Training Techniques | ✅ | ||
| Understanding Transformers... (beyond the Math) | ✅ | ||
| SFT, RL, and On-Policy Distillation Through a Distributional Lens | ✅ |
Multi-modality (Vision)
| Format | Resource | Length | Progress |
|---|---|---|---|
| DeepLearning.AI | How Diffusion Models Work | 46m | ✅ |
| Prompt Engineering for Vision Models | 1h32m | ⬜ | |
| Building Multimodal Search and RAG | 1h32m | ✅ | |
| Youtube | Lesson 9A 2022 - Stable Diffusion deep dive | 41m | ✅ |
| AI Visions Live | Merve Noyan | Open-source Multimodality | 54m | ✅ | |
| Sander Dieleman - Generative modelling through iterative refinement | 1h4m | ✅ | |
| Article | Pinecone: Embedding Methods for Image Search | 0/8 | ⏳ |
| Understanding Multimodal LLMs | ✅ | ||
| Computer-Using Agent | ✅ | ||
| Flow Matching in 5 Minutes | ✅ | ||
| Understanding Patch Embeddings for Vision Transformers (ViT) | ✅ | ||
| Diffusion models are autoencoders | ✅ | ||
| Diffusion Language Models | ✅ | ||
| Guidance: a cheat code for diffusion models | ✅ | ||
| Perspectives on diffusion | ✅ | ||
| The geometry of diffusion guidance | ✅ | ||
| Diffusion is spectral autoregression | ✅ | ||
| Generative modelling in latent space | ✅ |
Multi-modality (Audio)
| Format | Resource | Length | Progress |
|---|---|---|---|
| Article | Speech AI models: an introduction | ✅ | |
| Voice AI & Voice Agents - An Illustrated Primer | ⬜ | ||
| Neural audio codecs: how to get audio into LLMs | ⬜ |
Quantization
| Format | Resource | Length | Progress |
|---|---|---|---|
| DeepLearning.AI | Quantization Fundamentals with Hugging Face | 1h14m | ✅ |
| Quantization in Depth | 2h20m | ⬜ | |
| Introduction to On-Device AI | 1h19m | ⬜ | |
| Youtube | CMU Advanced NLP Fall 2024 (11): Distillation, Quantization, and Pruning | 1h4m | ⬜ |
| Article | A Visual Guide to Quantization | ⬜ | |
| QLoRA and 4-bit Quantization | ⬜ | ||
| Understanding AI/LLM Quantisation Through Interactive Visualisations | ⬜ | ||
| LLM.int8() and Emergent Features | ⬜ | ||
| Quantization from the ground up | ⬜ |
Systems Engineering
| Format | Resource | Length | Progress |
|---|---|---|---|
| Book | Inference Engineering | 275 pgs | ✅ |
| Programming Massively Parallel Processors: A Hands-on Approach | Ch. 2 | ⏳ | |
| The Algebra of Speed | ⬜ | ||
| DeepLearning.AI | Efficiently Serving LLMs | 2h40m | ✅ |
| Udacity | Intro to Parallel Programming | 299/458 | ⏳ |
| Youtube | GPU Puzzles: Let's Play | 1h15m | ⬜ |
| Deploying Fine-Tuned Models | 2h28m | ✅ | |
| Building Machine Learning Systems for a Trillion Trillion Floating Point Operations | 1h | ✅ | |
| How GPT, Claude, and Gemini are actually trained and served – Reiner Pope | 2h13m | ✅ | |
| Chip design from the bottom up – Reiner Pope | 1h20m | ✅ | |
| Street Fighting Transformers | 25m | ✅ | |
| Slaying OOMs with PyTorch FSDP and torchao | 49m | ✅ | |
| Distributed Training with PyTorch: complete tutorial with cloud infrastructure and code | 1h12m | ✅ | |
| How DDP works || Distributed Data Parallel | 3m | ✅ | |
| FSDP Explained | 11m | ✅ | |
| Lecture 48: The Ultra Scale Playbook | 44m/3h3m | ⏳ | |
| Invited Talk: PyTorch Distributed (DDP, RPC) - By Facebook Research Scientist Shen Li | 1h7m | ✅ | |
| Unit 9 | Techniques for Speeding Up Model Training | 12/12 | ✅ | |
| Optimizing Model Training End-to-End: A Tiny MoE Case Study Lambda | 19m | ✅ | |
| Article | How to make LLMs go fast | ✅ | |
| In the Fast Lane! Speculative Decoding - 10x Larger Model, No Extra Cost | ⬜ | ||
| Accelerating Generative AI with PyTorch II: GPT, Fast | ⬜ | ||
| Harmonizing Multi-GPUs: Efficient Scaling of LLM Inference | ⬜ | ||
| Multi-Query Attention is All You Need | ⬜ | ||
| Transformers Inference Optimization Toolset | ⬜ | ||
| LLM Inference Series: 3. KV caching explained | ⬜ | ||
| LLM Inference Series: 4. KV caching, a deeper look | ⬜ | ||
| LLM Inference Series: 5. Dissecting model performance | ⬜ | ||
| Transformer Inference Arithmetic | ✅ | ||
| llama.cpp guide - Running LLMs locally, on any hardware, from scratch | ✅ | ||
| Domain specific architectures for AI inference | ⬜ | ||
| SBTB 2023: Charles Frye, Parallel Processors: Past & Future Connections Between LLMs and OS Kernels | 32m | ✅ | |
| Compiling ML models to C for fun | ⬜ | ||
| How to Optimize a CUDA Matmul Kernel for cuBLAS-like Performance: a Worklog | ⬜ | ||
| Inside vLLM: Anatomy of a High-Throughput LLM Inference System | ⬜ | ||
| Understanding LLM Inference Engines: Inside Nano-vLLM (Part 1) | ⬜ | ||
| Understanding LLM Inference Engines: Inside Nano-vLLM (Part 2) | ⬜ | ||
| Defeating Nondeterminism in LLM Inference | ✅ | ||
| Prompt caching: 10x cheaper LLM tokens, but how? | ⬜ | ||
| Attention Mechanism: From Math to GPU | ⬜ | ||
| Making Deep Learning Go Brrrr From First Principles | ✅ | ||
| What Shapes Do Matrix Multiplications Like? | ✅ | ||
| Strangely, Matrix Multiplications on GPUs Run Faster When Given "Predictable" Data! | ✅ | ||
| A Short Guide to PyTorch DDP | ✅ | ||
| Scaling Deep Learning with PyTorch: Multi-Node and Multi-GPU Training Explained (with Code) | ✅ | ||
| Accelerating PyTorch Model Training | ✅ | ||
| Meet Horovod: Uber’s Open Source Distributed Deep Learning Framework for TensorFlow | ✅ | ||
| Distributed data parallel training in Pytorch | ✅ | ||
| Training on Multiple GPUs | ✅ |
Finetuning and Distillation
| Format | Resource | Length | Progress |
|---|---|---|---|
| DeepLearning.AI | Finetuning Large Language Models | 1h35m | ✅ |
| OpenAI | GPT-4o mini Fine-Tuning Build Hour | ✅ | |
| Distillation Build Hour | ✅ | ||
| Youtube | Fine-Tuning with Axolotl | 2h10m | ✅ |
| Creating, Curating, and Cleaning Data for LLMs | 54m | ✅ | |
| Best Practices For Fine Tuning Mistral | 23m | ✅ | |
| Fine Tuning OpenAI Models - Best Practices | 49m | ✅ | |
| When and Why to Fine Tune an LLM | 1h56m | ✅ | |
| Napkin Math For Fine Tuning Pt. 1 w/Johno Whitaker | 57m | ✅ | |
| Napkin Math For Fine Tuning Pt. 2 w/Johno Whitaker | 1h4m | ✅ | |
| Fine Tuning LLMs for Function Calling w/Pawel Garback | 1h32m | ✅ | |
| From Prompt to Model: Fine-tuning when you've already deployed LLMs in prod w/Kyle Corbitt | 32m | ✅ | |
| Why Fine Tuning is Dead w/Emmanuel Ameisen | 50m | ✅ | |
| Article | Tokenization Gotchas | ⬜ | |
| Practical Tips for Finetuning LLMs Using LoRA (Low-Rank Adaptation) | ⬜ | ||
| How to Generate and Use Synthetic Data for Finetuning | ✅ |
AI Engineering
LLM System Design
| Format | Resource | Length | Progress |
|---|---|---|---|
| Book | AI Engineering: Building Applications with Foundation Models | 400 pgs | ✅ |
| Course | LLM Bootcamp - Spring 2023 | ✅ | |
| DeepLearning.AI | Building Systems with the ChatGPT API | 1h55m | ✅ |
| Building Generative AI Applications with Gradio | 59m | ✅ | |
| LLMOps | 1h31m | ⬜ | |
| Youtube | A Survey of Techniques for Maximizing LLM Performance | 45m | ✅ |
| Building Blocks for LLM Systems & Products: Eugene Yan | 17m | ✅ | |
| Building LLM Applications | 0/8 | ⏳ | |
| Getting the Most Out of Your LLM Experiments | 48m | ✅ | |
| Article | What We’ve Learned From A Year of Building with LLMs | ✅ | |
| Data Flywheels for LLM Applications | ✅ | ||
| Emerging Architectures for LLM Applications | ✅ | ||
| Patterns for Building LLM-based Systems & Products | ✅ | ||
| LLM From the Trenches: 10 Lessons Learned Operationalizing Models at GoDaddy | ✅ | ||
| Emerging UX Patterns for Generative AI Apps & Copilots | ✅ | ||
| Pushing ChatGPT's Structured Data Support To Its Limits | ✅ | ||
| GPTed: using GPT-3 for semantic prose-checking | ✅ | ||
| Don't worry about LLMs | ✅ | ||
| Data acquisition strategies for AI-first start-ups | ✅ | ||
| All about synthetic data generation | ✅ | ||
| The Synthetic Data Playbook: Generating Trillions of the Finest Tokens | ⬜ | ||
| Reverse-engineering Claude's generative UI - then building it for the terminal | ✅ | ||
| Redpoint 2026 Market Update | ✅ | ||
| Designing synthetic datasets for the real world: Mechanism design and reasoning from first principles | ✅ |
Evals and Guardrails
| Format | Resource | Length | Progress |
|---|---|---|---|
| DeepLearning.AI | Automated Testing for LLMOps | 1h12m | ✅ |
| Red Teaming LLM Applications | 1h29m | ✅ | |
| Evaluating and Debugging Generative AI Models Using Weights and Biases | 1h | ⬜ | |
| Quality and Safety for LLM Applications | 1h59m | ⬜ | |
| Youtube | OpenAI: Evals Build Hour | ✅ | |
| Instrumenting & Evaluating LLMs | 2hr33m | ✅ | |
| LLM Eval For Text2SQL | 51m | ✅ | |
| A Deep Dive on LLM Evaluation | 49m | ✅ | |
| ARC Prize Foundation @ MIT 2025 | 8/8 | ✅ | |
| Article | Understanding the 4 Main Approaches to LLM Evaluation (From Scratch) | ✅ | |
| Your AI Product Needs Evals | ✅ | ||
| Task-Specific LLM Evals that Do & Don't Work | ✅ | ||
| Evaluation & Hallucination Detection for Abstractive Summaries | ✅ | ||
| Evaluating the Effectiveness of LLM-Evaluators (aka LLM-as-Judge) | ⬜ | ||
| Aligning LLM as judge with human evaluators | ✅ | ||
| Hard-Earned Lessons from 2 Years of Improving AI Applications | ✅ | ||
| Evaluating Long-Context Question & Answer Systems | ⬜ | ||
| LLM Evals FAQ | ✅ | ||
| Quantifying infrastructure noise in agentic coding evals | ✅ | ||
| How we optimized Dash's relevance judge with DSPy | ✅ | ||
| “It’s Hard to Eval” Is a Product Smell | ✅ |
Context Engineering
| Format | Resource | Length | Progress |
|---|---|---|---|
| Book | Prompt Engineering for LLMs | 250 pgs | ⬜ |
| Course | LLM Engineering - Structured Outputs | ✅ | |
| Anthropic | Anthropic Courses | ⬜ | |
| Anthropic: The Claude in Amazon Bedrock Course | ⬜ | ||
| DeepLearning.AI | Reasoning with o1 | 1h44m | ✅ |
| ChatGPT Prompt Engineering for Developers | 1h40m | ✅ | |
| Youtube | OpenAI: Reasoning with o1 Build Hour | ✅ | |
| Prompt Engineering Overview | 1h4m | ✅ | |
| Prompt Engineering Workshop | 1h | ✅ | |
| Context Engineering SF - August 20th, 2025 | 4/4 | ✅ | |
| Article | OpenAI Prompt Engineering | ✅ | |
| Prompting Fundamentals and How to Apply them Effectively | ✅ | ||
| How I came in first on ARC-AGI-Pub using Sonnet 3.5 with Evolutionary Test-time Compute | ✅ | ||
| Prompt Engineering(Liliang Weng) | ✅ | ||
| Prompt Engineering 201: Advanced methods and toolkits | ✅ | ||
| Optimizing LLMs for accuracy | ✅ | ||
| Primers • Prompt Engineering | ✅ | ||
| Guided text generation with Large Language Models | ✅ | ||
| Effective context engineering for AI agents | ✅ |
Information Retrieval / RAG
| Format | Resource | Length | Progress |
|---|---|---|---|
| Course | Course: Fullstack Retrieval | ⬜ | |
| Systematically Improving RAG | ⬜ | ||
| DeepLearning.AI | Building and Evaluating Advanced RAG Applications | 2h5m | ✅ |
| Vector Databases: from Embeddings to Applications | 1h5m | ✅ | |
| Advanced Retrieval for AI with Chroma | 1h2m | ✅ | |
| Prompt Compression and Query Optimization | 1h49m | ✅ | |
| Large Language Models with Semantic Search | 1h22m | ✅ | |
| Building Applications with Vector Databases | 1h23m | ✅ | |
| Knowledge Graphs for RAG | 2h4m | ⬜ | |
| Preprocessing Unstructured Data for LLM Applications | 1h22m | ⬜ | |
| Embedding Models: From Architecture to Implementation | 1h | ✅ | |
| Retrieval Optimization - From Tokenization to Vector Quantization | 1h43m | ✅ | |
| Youtube | Systematically improving RAG applications | 1h8m | ✅ |
| Back to Basics for RAG w/ Jo Bergum | 44m | ✅ | |
| Beyond the Basics of Retrieval for Augmenting Generation (w/ Ben Clavié) | 48m | ✅ | |
| RAG From Scratch | 14/14 | ✅ | |
| CMU Advanced NLP Fall 2024 (10): Retrieval and RAG | 1h17m | ✅ | |
| What You See Is What You Search: Vision Language Models for PDF Retrieval [Jo Bergum] | 44m | ✅ | |
| Graph Databases: When to Use Them (And When to Run Away) | 33m | ✅ | |
| Personalization in the Era of LLMs | 20m | ✅ | |
| Intelligence Efficiency, Ben Geist | Compile 26 | 14m | ✅ | |
| The Memory Problem, Baseten | Compile 26 | 13m | ✅ | |
| Article | Pretrained Transformer Language Models for Search - part 1 | ✅ | |
| Pretrained Transformer Language Models for Search - part 2 | ✅ | ||
| Pretrained Transformer Language Models for Search - part 3 | ✅ | ||
| Pretrained Transformer Language Models for Search - part 4 | ✅ | ||
| How not to use BERT for Document Ranking | ✅ | ||
| Understanding LanceDB's IVF-PQ index | ✅ | ||
| A little pooling goes a long way for multi-vector representations | ✅ | ||
| Levels of Complexity: RAG Applications | ✅ | ||
| Systematically Improving Your RAG | ✅ | ||
| Stop using LGTM@Few as a metric (Better RAG) | ✅ | ||
| Low-Hanging Fruit for RAG Search | ✅ | ||
| What AI Engineers Should Know about Search | ✅ | ||
| Evaluating Chunking Strategies for Retrieval | ✅ | ||
| Sentence Embeddings. Introduction to Sentence Embeddings | ✅ | ||
| LambdaMART in Depth | ⬜ | ||
| Guided Generation with Outlines | ✅ | ||
| RAG tricks from the trenches | ✅ | ||
| Understanding the BM25 full text search algorithm | ⬜ | ||
| Arxiv: Ragas: Automated Evaluation of Retrieval Augmented Generation | ✅ | ||
| Pinecone: Vector Databases in Production for Busy Engineers | ✅ | ||
| Pinecone: Retrieval Augmented Generation | ✅ | ||
| Pinecone: Faiss: The Missing Manual | ✅ | ||
| Pinecone: Natural Language Processing for Semantic Search | 0/13 | ⏳ | |
| Guidance: Token Healing | ⬜ | ||
| Deep Research System Card | ✅ | ||
| Exploring OpenAI's deep research API model o4-mini-deep-research | ✅ | ||
| Introduction to deep research in the OpenAI API | ✅ | ||
| Lessons from building the best Deep Research (and how you can build better agents) | ✅ |
Agentic Engineering
| Format | Resource | Length | Progress |
|---|---|---|---|
| DeepLearning.AI | Agentic AI with Andrew Ng | 9h55m | ✅ |
| Building Agentic RAG with LlamaIndex | 54m | ✅ | |
| Multi AI Agent Systems with crewAI | 3h1m | ✅ | |
| Building Towards Computer Use with Anthropic | 1h47m | ✅ | |
| Pydantic for LLM Workflows | 1h50m | ✅ | |
| Practical Multi AI Agents and Advanced Use Cases with crewAI | 2h49m | ⬜ | |
| LLMs as Operating Systems: Agent Memory | 1h | ✅ | |
| Serverless Agentic Workflows with Amazon Bedrock | 1h | ⬜ | |
| AI Agentic Design Patterns with AutoGen | 1h | ⬜ | |
| AI Agents in LangGraph | 1h42m | ⬜ | |
| Building Your Own Database Agent | 1h8m | ⬜ | |
| Function-Calling and Data Extraction with LLMs | 59m | ✅ | |
| Evaluating AI Agents | 2h16m | ✅ | |
| Building AI Browser Agents | 1h5m | ⬜ | |
| Course | Berkeley: CS294/194-196 Large Language Model Agents | 0/14 | ⬜ |
| Huggingface: Agents Course | Unit 1 | ⏳ | |
| Berkeley: Advanced LLM Agents MOOC | 0/12 | ⬜ | |
| Youtube | OpenAI: Assistants & Agents Build Hour | ✅ | |
| OpenAI: Function Calling Build Hour | ✅ | ||
| How to Evaluate Agents: Galileo’s Agentic Evaluations in Action | 16m | ✅ | |
| Agent Response | LangSmith Evaluation - Part 24 | 14m | ✅ | |
| Single Step | LangSmith Evaluation - Part 25 | 6m | ✅ | |
| Agent Trajectory | LangSmith Evaluation - Part 26 | 8m | ✅ | |
| Evaluating Agents and Assistants: The AI Conference | 24m | ✅ | |
| How to Build, Evaluate, and Iterate on LLM Agents | 1h2m | ✅ | |
| How Claude Code Works - Jared Zoneraich, PromptLayer | 1h5m | ✅ | |
| Build Agents That Run for Hours (Without Losing the Plot) — Ash Prabaker & Andrew Wilson, Anthropic | 1h15m | ✅ | |
| Article | Tool Invocation - Demonstrating the Marvel of GPT's Flexibility | ✅ | |
| Introducing smolagents, a simple library to build agents | ✅ | ||
| What Problem Does The Model Context Protocol Solve? | ✅ | ||
| Don’t Build Multi-Agents | ✅ | ||
| Coding Agents 101: The Art of Actually Getting Things Done | ✅ | ||
| What makes Claude Code so damn good (and how to recreate that magic in your agent)!? | ✅ | ||
| Anthropic: Building effective agents | ✅ | ||
| Anthropic: Building Effective Agents Cookbook | ✅ | ||
| Anthropic: Writing effective tools for agents — with agents | ✅ | ||
| How System Prompts Define Agent Behavior | ✅ | ||
| Components of A Coding Agent | ✅ |
Traditional Machine Learning
| Format | Resource | Length | Progress |
|---|---|---|---|
| Book | Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow, 2nd Edition | 848 pgs | ⬜ |
| A Machine Learning Primer | 100 pgs | ✅ | |
| Grokking Machine Learning | 320 pgs | ✅ | |
| The StatQuest Illustrated Guide To Machine Learning | 304 pgs | ✅ | |
| Datacamp | Ensemble Methods in Python | 4h | ✅ |
| Extreme Gradient Boosting with XGBoost | 4h | ⬜ | |
| Clustering Methods with SciPy | 4h | ✅ | |
| Unsupervised Learning in Python | 4h | ✅ | |
| Supervised Learning with scikit-learn | 4h | ✅ | |
| Machine Learning with Tree-Based Models in Python | 5h | ✅ | |
| Linear Classifiers in Python | 4h | ✅ | |
| Model Validation in Python | 4h | ✅ | |
| Hyperparameter Tuning in Python | 4h | ✅ | |
| HR Analytics in Python: Predicting Employee Churn | 4h | ✅ | |
| Predicting Customer Churn in Python | 4h | ✅ | |
| Dimensionality Reduction in Python | 4h | ✅ | |
| Preprocessing for Machine Learning in Python | 4h | ✅ | |
| Data Types for Data Science | 4h | ✅ | |
| Cleaning Data in Python | 4h | ✅ | |
| Feature Engineering for Machine Learning in Python | 4h | ✅ | |
| Predicting CTR with Machine Learning in Python | 4h | ✅ | |
| Intro to Financial Concepts using Python | 4h | ✅ | |
| Fraud Detection in Python | 4h | ✅ | |
| Segmentation and Clustering | 5/5 | ✅ | |
| Intro to Python for Data Science | 4h | ✅ | |
| edX | Implementing Predictive Analytics with Spark in Azure HDInsight | ✅ | |
| Article | The wrong batch size is all it takes | ✅ | |
| A Gentle Introduction to Expectation-Maximization (EM Algorithm) | ✅ | ||
| How to Use Out-of-Fold Predictions in Machine Learning | ✅ | ||
| Stacking and Blending — An Intuitive Explanation | ✅ | ||
| Youtube | A walk through the isolation forest | 23m | ✅ |
| Complete Anomaly Detection Tutorials Machine Learning And Its Types With Implementation | 36m | ✅ |
Deep Learning
| Format | Resource | Length | Progress |
|---|---|---|---|
| Book | Make Your Own Neural Network | 226 pgs | ✅ |
| Fast.ai | Practical Deep Learning for Coder (Part 1) | ✅ | |
| Practical Deep Learning for Coder (Part 2) | 9, 13,14,17,18(48:10),19 | ⬜ | |
| Datacamp | Convolutional Neural Networks for Image Processing | 4h | ✅ |
| Karpathy | Neural Networks: Zero to Hero | ✅ | |
| Article | An overview of gradient descent optimization algorithms | ✅ | |
| Things that confused me about cross-entropy | ✅ | ||
| Why is the ReLU function not differentiable at x=0? | ✅ | ||
| Are CNNs invariant to translation, rotation, and scaling? | ✅ | ||
| How to Control the Stability of Training Neural Networks With the Batch Size | ✅ | ||
| A Visual Guide to Learning Rate Schedulers in PyTorch | ✅ | ||
| CS231N Notes on Optimization | ✅ |
Technical Skills (Libraries/Frameworks/Tools)
CSS
| Format | Resource | Length | Progress |
|---|---|---|---|
| Pluralsight | CSS Positioning | ✅ | |
| Introduction to CSS | ✅ | ||
| CSS: Specificity, the Box Model, and Best Practices | ✅ | ||
| CSS: Using Flexbox for Layout | ✅ | ||
| Code School: Blasting Off with Bootstrap | ✅ | ||
| Codecademy | Learn SASS | 3h | ✅ |
| Course | CSS for Javascript Developers | ✅ |
Django
| Format | Resource | Length | Progress |
|---|---|---|---|
| Article | Django, HTMX and Alpine.js: Modern websites, JavaScript optional | ✅ |
HTML
| Format | Resource | Length | Progress |
|---|---|---|---|
| Codecademy | Learn HTML | 7h | ✅ |
| Make a website | 4h | ✅ | |
| Article | Alternative Text | ⬜ |
Langchain
| Format | Resource | Length | Progress |
|---|---|---|---|
| Course | Pinecone: LangChain AI Handbook | 0/11 | ⏳ |
| DeepLearning.AI | LangChain for LLM Application Development | 1h48m | ⬜ |
| LangChain: Chat with Your Data | 1h18m | ⬜ |
JavaScript
| Format | Resource | Length | Progress |
|---|---|---|---|
| Codecademy | Learn JavaScript | ✅ | |
| Jquery Track | 4h | ✅ | |
| Udacity | ES6 - JavaScript Improved | 4/4 | ✅ |
| Intro to Javascript | 7/7 | ✅ | |
| Object Oriented JS 1 | 8/8 | ✅ | |
| Object Oriented JS 2 | 4/4 | ✅ | |
| Udemy | Understanding Typescript | ✅ |
Matplotlib
| Format | Resource | Length | Progress |
|---|---|---|---|
| Datacamp | Introduction to Seaborn | 4h | ✅ |
| Introduction to Matplotlib | 4h | ✅ |
MLFlow
| Format | Resource | Length | Progress |
|---|---|---|---|
| Datacamp | Introduction to MLFlow | 4h | ✅ |
Numpy
| Format | Resource | Length | Progress |
|---|---|---|---|
| Youtube | Numpy Array Broadcasting In Python Explained | 13m | ✅ |
Nexxt.JS
| Format | Resource | Length | Progress |
|---|---|---|---|
| Docs | Start building with Next.js | ⬜ |
Pandas
| Format | Resource | Length | Progress |
|---|---|---|---|
| Datacamp | Pandas Foundations | 4h | ✅ |
| Pandas Joins for Spreadsheet Users | 4h | ✅ | |
| Manipulating DataFrames with pandas | 4h | ✅ | |
| Merging DataFrames with pandas | 4h | ✅ | |
| Data Manipulation with pandas | 4h | ✅ | |
| Optimizing Python Code with pandas | 4h | ✅ | |
| Streamlined Data Ingestion with pandas | 4h | ✅ | |
| Analyzing Marketing Campaigns with pandas | 4h | ✅ | |
| Analyzing Police Activity with pandas | 4h | ✅ |
PyTorch
| Format | Resource | Length | Progress |
|---|---|---|---|
| Course | Deeplizard: Neural Network Programming - Deep Learning with PyTorch | 20/20 | ✅ |
| Datacamp | Introduction to Deep Learning with PyTorch | 4h | ✅ |
| Intermediate Deep Learning with PyTorch | 4h | ✅ | |
| Deep Learning for Text with PyTorch | 4h | ⬜ | |
| Deep Learning for Images with PyTorch | 4h | ⬜ | |
| Article | PyTorch internals | ⬜ | |
| Taking PyTorch For Granted | ⬜ |
ReactJS
| Format | Resource | Length | Progress |
|---|---|---|---|
| Codecademy | Learn ReactJS: Part I | 13h | ✅ |
| Learn ReactJS: Part II | 3h | ✅ | |
| Course | NexxtJS: React Foundations | ⬜ |
Spacy
| Format | Resource | Length | Progress |
|---|---|---|---|
| Datacamp | Advanced NLP with spaCy | 4h | ✅ |
Tensorflow & Keras
| Format | Resource | Length | Progress |
|---|---|---|---|
| Course | Deeplizard: Keras - Python Deep Learning Neural Network API | 21/21 | ✅ |
| Datacamp | Introduction to TensorFlow in Python | 4h | ✅ |
| Deep Learning in Python | 4h | ✅ | |
| Introduction to Deep Learning with Keras | 4h | ✅ | |
| Advanced Deep Learning with Keras | 4h | ✅ |
Repo liên quan
Hugging Face Transformers (huggingface/transformers) is a Python library that centralizes model definitions for text, computer vision, audio, video, and multimodal machine learning, covering both inference and training. The README describes it as a pivot point compatible with training frameworks such as Axolotl, DeepSpeed, and PyTorch-Lightning, and inference engines such as vLLM, SGLang, and TGI, with more than 1M+ model checkpoints listed on the Hugging Face Hub.
LLMs-from-scratch is Sebastian Raschka's GitHub companion to the Manning book Build a Large Language Model (From Scratch), ISBN 9781633437166. It walks through building, pretraining, and finetuning a GPT-style model in PyTorch across seven chapters and five appendices, plus a growing set of bonus notebooks covering newer open architectures.
What Hacker News commenters said about TensorFlow's installs, sprawl and 2.x migration, set against the repo's 201,431 stars and 2026-11-10 push.
PyTorch is an open-source Python package hosted at pytorch/pytorch on GitHub, built around two core capabilities: NumPy-like Tensor computation with GPU acceleration, and deep neural networks built through a tape-based autograd system. Its components include torch.nn for network layers, torch.jit for compiling models into TorchScript, and torch.multiprocessing for sharing Tensors across processes, with build support for CUDA, AMD ROCm, and Intel GPU backends documented in the README.
Trả lời nhanh
amitness/learning có những chủ đề gì?
GitHub topics của amitness/learning: "deep-learning", "generative-ai", "learning-resources", "llms", "machine-learning", "nlp", "python". TopGit xếp repo vào nhóm AI Tools.
amitness/learning có phải mã nguồn mở không?
Có — amitness/learning phát hành theo license MIT, nghĩa là mã nguồn mở để đọc, fork và (tùy license) tái sử dụng. Mã: github.com/amitness/learning.
amitness/learning có trang demo không?
Dự án có trang chủ ở https://twitter.com/amitness. Tab "Readme" ở trang này thường có ảnh chụp và hướng dẫn bắt đầu nhanh.
amitness/learning còn đang phát triển không?
Commit gần nhất trên amitness/learning là 4 tháng trước (theo timestamp GitHub). Repo có 873 fork — một chỉ báo về mức độ quan tâm của cộng đồng.
amitness/learning dùng license gì?
amitness/learning phát hành theo license MIT. Nên mở file LICENSE trên GitHub để xác nhận — license metadata đôi khi lệch với thực tế dự án.
Đọc thêm về amitness/learning ở đâu?
Trang TopGit này là một snapshot — tab "Readme" hiển thị nguyên văn README của repo (đã bỏ link, giữ ảnh). Repo GitHub ở github.com/amitness/learning là nguồn chính thức.
Đọc đầy đủ README ở tab phía trên.
Vẫn đang phân vân về learning?
Một cú bấm sẽ gửi câu hỏi kèm trang này cho AI — xem AI nói gì về learning.