Main Page
Welcome to Leeroopedia
Your ML & Data Knowledge Wiki. Best practices and expert-level knowledge for Machine Learning and Data Engineering, covering 1000+ frameworks and libraries from training to deployment.
Browse implementation patterns, configuration guides, debugging heuristics, and battle-tested defaults for frameworks like vLLM, DeepSpeed, Megatron-LM, FlashAttention, Triton, Unsloth, LangChain, and many more. Every page is structured so both humans and AI agents can find what they need fast.
Connect your AI coding agent. Plug Leeroopedia into your favorite coding agent, and let it build robust AI/ML systems autonomously:
- SuperML plugin — converts your AI coding agent into an expert ML engineer with agentic memory
- Leeroopedia MCP — search over best-practices and skills of ML/AI
- Kapso — experimentation platform for autonomous AI/ML software building
Browse by Category
| Category | Description | Browse |
|---|---|---|
| Workflows | Step-by-step processes and procedures | Browse All |
| Principles | Core ideas and foundational knowledge | Browse All |
| Implementations | Code-level details and modules | Browse All |
| Heuristics | Best practices and guidelines | Browse All |
| Environments | Setup and configuration guides | Browse All |
Explore Pages
Workflows
- Workflow:AnswerDotAI RAGatouille In Memory Retrieval
- Workflow:Cleanlab Cleanlab Multiannotator Consensus
- Workflow:Fede1024 Rust rdkafka At Least Once Processing
- Workflow:Dotnet Machinelearning ONNX Model Scoring
- Workflow:Ggml org Llama cpp LoRA Adapter Workflow
- Workflow:VainF Torch Pruning Vision Transformer Pruning
- Workflow:TA Lib Ta lib python Candlestick Pattern Recognition
- Workflow:Langchain ai Langgraph Functional API Workflow
- Workflow:Apache Flink Hybrid Source Switching
- Workflow:Openai Evals Creating a model graded eval
Principles
- Principle:FlowiseAI Flowise Dependency Management
- Principle:Mistralai Client python Function Dispatch
- Principle:Ggml org Llama cpp Tensor Debugging
- Principle:Rapidsai Cuml Clustering Algorithm Configuration
- Principle:Pyro ppl Pyro Online Statistics
- Principle:Huggingface Datatrove Pipeline Failure Diagnosis
- Principle:Togethercomputer Together python Chat Completion Response
- Principle:Duckdb Duckdb Prerequisite Code Generation
- Principle:Huggingface Peft Cosine Similarity Contrastive Training
- Principle:ARISE Initiative Robosuite Visual Feedback Wrapping
Implementations
- Implementation:CARLA simulator Carla Signal
- Implementation:Apache Airflow Provider Yaml Schema
- Implementation:Helicone Helicone ClickhousePriceCalc
- Implementation:Sgl project Sglang CPU QKV Projection
- Implementation:Ggml org Llama cpp Finetune
- Implementation:Bitsandbytes foundation Bitsandbytes PagedAdamW8bit
- Implementation:Risingwavelabs Risingwave MongoDbValidator
- Implementation:Ucbepic Docetl ColumnDialog
- Implementation:Pyro ppl Pyro SIR Models
- Implementation:Turboderp org Exllamav2 Ext QAttn
Heuristics
- Heuristic:Scikit learn contrib Imbalanced learn KNeighbors Selection Tips
- Heuristic:PeterL1n BackgroundMattingV2 Checkpoint Interval Tuning
- Heuristic:Anthropics Anthropic sdk python Retry Backoff Strategy
- Heuristic:Langchain ai Langchain Warning Deprecated AnthropicLLM
- Heuristic:Hpcaitech ColossalAI CUDA Device Max Connections Tip
- Heuristic:Haifengl Smile Quarkus Async Context Handling
- Heuristic:Onnx Onnx Warning Deprecated InlineSelectedFunctions
- Heuristic:Microsoft Autogen Parallel Tool Call Safety
- Heuristic:Vespa engine Vespa Config Polling Timeout Tuning
- Heuristic:Openai CLIP CLIP Normalization Constants
Environments
- Environment:TobikoData Sqlmesh Python Runtime
- Environment:Bitsandbytes foundation Bitsandbytes Build From Source Environment
- Environment:Eventual Inc Daft Python PyArrow Core
- Environment:Microsoft Playwright Node Runtime Environment
- Environment:Sktime Pytorch forecasting Optuna Tuning Dependencies
- Environment:Unslothai Unsloth CUDA BitsAndBytes
- Environment:Helicone Helicone Docker Compose Infrastructure
- Environment:Cypress io Cypress Browser Requirements
- Environment:Datajuicer Data juicer GPU CUDA Environment
- Environment:Dagster io Dagster Container Resource Monitoring