ai-pipelines

All-in-One Development Tool based on PaddlePaddle

PostgreSQL in-database durable execution

An open-source data logging library for machine learning models and data pipelines. 📚 Provides visibility into data quality & model performance over time. 🛡️ Supports privacy-preserving data collection, ensuring safety & robustness. 📈

VexJoy AI Agent with Jev Intelligent Routing - /do routes plain-English requests to the right specialist agent and gates the work with reviews, tests, and a learning loop.

Profile and monitor your ML data pipeline end-to-end

Kedro Plugin to support running workflows on Kubeflow Pipelines

An open library of robotics data transformations

Agentic AI refers to AI systems capable of autonomous decision-making, planning, and executing tasks based on goals—acting like intelligent agents. These systems combine LLMs with tools, memory, and feedback loops to complete complex workflows with minimal human input.

LLM pipelines for the Islam West Africa Collection: OCR, HTR, transcription, NER, summarization and sentiment analysis, written back to Omeka S

NVIDIA Speech Artificial Intelligence. Speech AI Summit 2022.

A pixel-art game engine with built-in editor, AI asset generation, and algorithmic chiptune music.

Mira is an assistive AI agent that helps a victim of non-consensual intimate imagery (NCII / sexual deepfakes) get the content removed under EU law — without ever having to look at the material, hunt down the host, or write a single line of legal text. Realized in raise hackathon 2026

YES AI Studio — Agentic Multimodal Creator (Capstone Edition). A fully agentic AI system that combines multimodal generation, intelligent workflows, autonomous task execution, and content creation pipelines. Designed as a capstone-level project to showcase advanced AI engineering using Python, vision models, NLP, and tool-based agent execution.

Full-stack web app wrapping a bounded, testable LLM pipeline for open-ended qualitative analysis: staged contracts between components, empirical calibration to reduce output variance, and a client-safe boundary enforced through to the human-review UI. Modeled on an existing DEI consultancy.

Declarative Self-Improving Framework for Auto-Prompt Optimization and LLM Program Compilation.

A production-ready .NET sample for reliable batch AI inference with Microsoft Foundry, featuring idempotency, checkpoint-based recovery, bounded retries, failure quarantine, and replayable outputs.

Multi-LLM pipeline for validated extraction and structuring of species-level botanical and ecological knowledge.

A command-line tool that runs a series of interconnected AI prompts to automate complex writing and analysis tasks.

🚀 Enhance workflows with a high-performance orchestration engine for routing tasks across LLMs, SLMs, and custom agents efficiently.

Observability for multi-step AI pipelines: traces each step as OpenTelemetry-shaped spans, pinpoints the first failing step (root cause, propagation-aware), and grows an eval dataset from flagged failures. Mini LangSmith.