The easiest way to serve AI apps and models - Build Model Inference APIs, Job queues, LLM apps, Multi-model pipelines, and more!

An AI-powered data science team of agents to help you perform common data science tasks 10X faster.

Kafka variant of the MLOps Level 1 stack

Scaffolding for serving ml model APIs using FastAPI

Missing Data Doctor is a diagnostic and treatment toolkit for missing values in machine learning datasets. It profiles missingness patterns, visualizes gaps, applies multiple imputation strategies, and evaluates their impact on model performance. Includes automated plots, metrics, and a full HTML report.

A playground for building and serving Retrieval-Augmented Generation (RAG) systems using best practices in MLOps and LLMOps, with open-source tools.

An easy-to-use tool for making web service with API from your own Python functions.

Crack SWE (ML) / DS MAANG Interviews

A complete production-ready MLOps framework with built-in distributed training, monitoring, and CI/CD. Deploy ML models to production with confidence using our battle-tested infrastructure.

All About ML Engineering

End-to-end deep learning system for sugarcane leaf disease detection using PyTorch, ResNet-50, FastAPI, and Docker.

Incremental learning with CatBoost and Ray for scalable training, tuning, and serving of large ML models

A task queue for serving machine learning models to a website -- RabbitMQ, Celery, all the good stuff.

🖼️ Extract text from images locally using Ollama's LLMs—100% free, offline, and private. No API keys or cloud costs necessary.

Benchmark pédagogique MLOps : Scikit-Learn vs XGBoost vs PyTorch sur données tabulaires

PyTorch computer-vision CLI tool for metadata-conditioned illumination normalization in dermatology-style images

QueryInsights is a learning-to-rank system that demonstrates how search results can be ordered using machine learning. It compares logistic regression and XGBoost models, evaluates ranking quality with NDCG@10, and simulates potential click-through improvements using an offline A/B framework.

The attenuator is based on a half waveplate and a polarizer: while the polarizer is fixed and transmits the p-polarization, the waveplate is housed in a motor-controlled 360° rotation mount and allows to continuously change the input linear polarization.

End-to-end ML system for ETA prediction with FastAPI backend, Docker deployment, Streamlit UI, authentication, and analytics dashboard.