GitHub-проекты

sigoptsklearn

seqlearn

scikit-feature

Data Science Degree @ Berkeley

Coursera Big Data Specialization

A hands-on course to train and deploy a serverless API that predicts crypto prices.

Open Source Society University

A straightforward method for training your LLM, from downloading data to generating text.

Realtime deployment Tutorial on Python time-series model deployment.

Machine Learning, Data Science and Deep Learning with Python

DataCamp Cheatsheets Cheatsheets for data science.

A weekly data project aimed at the R ecosystem.

Daily-updated skill and CLI for deterministic retrieval across arXiv, PubMed/PMC, and supported US policy corpora.

Open-source Python library and MCP server to benchmark document chunking strategies for RAG, score retrieval quality, and recommend configurations for a corpus.

Open-source LLM and agent evaluation framework with 50+ metrics, LLM-as-Judge augmentation, and guardrail scanners (jailbreak, PII, prompt-injection). Useful for scoring RAG outputs, agent trajectories, and function-calling behavior in data-science workflows.

Local AI agent for generating publication-ready scientific papers with real arXiv citations, IMRaD structure, and tribunal scoring. Runs 100% offline via Ollama with 4B-9B models. MIT licensed. HuggingFace

AI crypto trading framework using LightGBM + XGBoost ensemble with 72 ML features. 70.9% walk-forward validated accuracy on out-of-sample data. Supports Bybit and Binance. MIT licensed, available on PyPI.

Agent framework for chatting with data, turning natural language into SQL, transformation pipelines and visualizations. Outputs are declarative specs that can be inspected, edited, reopened in a notebook or composed into a dashboard.

Production-ready AI agent development kit for Rust with model-agnostic design (Gemini, OpenAI, Anthropic), multiple agent types (LLM, Graph, Workflow), MCP support, and built-in telemetry.

An open-source, workflow-first terminal coding agent with subagents, skills, sandboxed tools, and multiple model providers.

An open-source framework and registry for evaluating language models and systems.

DeepSeek's official setup guides for integrating its models with coding agents including Claude Code, Codex, Cline, OpenCode, and Pi.

A VS Code extension to view Weights & Biases experiments, logs, and artifacts within the IDE, eliminating the need to switch to the web UI and keeping data private.

A VS Code extension for viewing and exploring large machine learning datasets (CSV, JSON, Parquet, etc.) directly within the editor without VS Code crashing in a clean UI.

agentic AI operating system (h9y.ai) that replaces brittle/fragmented automations with long-lived, self-improving systems. Open-source, self-hosted/cloud, visual workflow, omni-channel, decentralized, extensible.

Open-source SDK for running LLMs and multimodal models on-device across iOS, Android, and cross-platform apps.

A curated collection of battle-tested tools, frameworks, and best practices for building, scaling, and monitoring production-grade Retrieval-Augmented Generation (RAG) systems. Covers frameworks, vector databases, retrieval & reranking, evaluation, observability, deployment, and security.

Curated list of top Hugging Face models for NLP, vision, and audio tasks with demos and benchmarks.

Godot 4.x asset that enables NPCs to interact with players using local LLMs for structured, offline-first learning conversations in games.

Chrome extension that uses local LLMs to assist with writing and drafting responses based on the context of your open tabs.

Open source Kubernetes-style control plane for deploying AI agents as distributed microservices, with built-in service discovery, durable workflows, and observability.

Visual AI agent workflow automation platform with local LLM integration. Build intelligent workflows using drag-and-drop, no cloud required.

Open-source runtime security scanner for AI agents. Detects prompt injection, jailbreak, PII leakage, memory poisoning, and tool misuse. Zero deps, MIT licensed.

Open-source CLI security scanner for agentic workflows. Scans your workflow’s source code, detects vulnerabilities, and generates an interactive visualization along with a detailed security report. Supports LangGraph, CrewAI, n8n, OpenAI Agents, and more.

Open-source LLM evaluation and red teaming framework. Test prompts, models, agents, and RAG pipelines. Run adversarial attacks (jailbreaks, prompt injection) and integrate security testing into CI/CD.

Ambrosia helps you clean up your LLM datasets using other LLMs.

Aqueduct enables you to easily define, run, and manage AI & ML tasks on any cloud infrastructure.

Ready to use deeplearning docker images.

Version and deploy your ML models following GitOps principles

ML powered analytics engine for outlier/anomaly detection and root cause analysis.

A library for doing continuous integration with ML projects. Use GitHub Actions & GitLab CI to train and evaluate models in production like environments and automatically generate visual reports with metrics and graphs in pull/merge requests. Framework & language agnostic.

A tool that allows the conversion of ML models into native code (Java, C, Python, Go, JavaScript, Visual Basic, C#, R, PowerShell, PHP, Dart) with zero dependencies.

Python tool to help you configure, organize, log and reproduce experiments. Like a notebook lab in the context of Chemistry/Biology. The community has built multiple add-ons leveraging the proposed standard.

a lightweight library to define data transformations as a directed-acyclic graph (DAG). It helps author reliable feature engineering and machine learning pipelines, and more.

Kedro is a data and development workflow framework that implements best practices for data pipelines with an eye towards productionizing machine learning models.

Python library for experiment metrics logging into simply formatted local files.

Data Science Version Control is an open-source version control system for machine learning projects with pipelines support. It makes ML projects reproducible and shareable.

Deepnote is a drop-in replacement for Jupyter with an AI-first design, sleek UI, new blocks, and native data integrations. Use Python, R, and SQL locally in your favorite IDE, then scale to Deepnote cloud for real-time collaboration, Deepnote agent, and deployable data apps.