Projects Recently Starred by Experts

Starred 10 hours ago by

Jeff Hammerbacher(Cofounder of Cloudera). Also starred by

Andrey Vasnetsov(Cofounder of Qdrant) and

Andre Zayarni(Cofounder of Qdrant).

swiftide by bosun-ai

0.8%

715

Rust library for building LLM applications

Fast data ingestion, transformation, and indexing for Retrieval Augmented Generation (RAG).
Flexible building blocks for creating agents; rapid development from concept to production.
Integrations with OpenAI, Groq, Redis, Qdrant, Ollama, FastEmbed-rs, and LanceDB.
Modular streaming indexing pipeline with async, parallel processing.

Created 2 years ago

Updated 3 days ago

Starred 10 hours ago by

Jeff Hammerbacher(Cofounder of Cloudera). Also starred by

Jason Huggins(Creator of Selenium),

Chip Huyen(Author of "AI Engineering", "Designing Machine Learning Systems"),

Georgios Konstantopoulos(CTO, General Partner at Paradigm), and

1 more.

rig by 0xPlaygrounds

0.7%

Rust library for building LLM-powered applications

Abstractions over LLM providers (OpenAI, Cohere) and vector stores (MongoDB, SQlite).
Supports LLM completion and embedding workflows with minimal boilerplate.
Vector store integrations: MongoDB, LanceDB, Neo4j, Qdrant, SQLite, SurrealDB.

Created 2 years ago

Updated 9 hours ago

Starred 15 hours ago by

Didier Lopes(Founder of OpenBB). Also starred by

Georgios Konstantopoulos(CTO, General Partner at Paradigm) and

Zhuohan Li(Coauthor of vLLM).

DeepSpec by deepseek-ai

65.5%

Codebase for training and evaluating speculative decoding

Train and evaluate draft models for speculative decoding algorithms.
Supports DSpark, DFlash, and Eagle3 algorithms with Qwen3 and Gemma models.
Full-stack utilities for data prep, training, and benchmark evaluation.
Optimizes inference speed by using draft models to predict token sequences.

Created 1 week ago

Updated 6 days ago

Starred 1 day ago by

Eric Zhu(Coauthor of AutoGen; Research Scientist at Microsoft Research). Also starred by

comfyanonymous(Author of ComfyUI; Cofounder of Comfy Org),

Joe Walnes(Head of Experimental Projects at Stripe),

Gregor Zunic(Cofounder of Browser Use), and

24 more.

ComfyUI by Comfy-Org

0.6%

120k

Visual AI engine for diffusion models, API, and backend

Node-based interface creates complex Stable Diffusion workflows without coding.
Supports SD1.x, SD2.x, SDXL, Stable Video Diffusion, SD3, and more.
Optimizes memory management for low VRAM GPUs; standalone VAEs/CLIP models.
Loads workflows from PNG/WebP/FLAC; saves/loads as JSON; works offline.

Created 3 years ago

Updated 15 hours ago

Starred 1 day ago by

Thiago Teixeira(Cofounder of Streamlit). Also starred by

Pietro Schirano(Founder of MagicPath),

Chip Huyen(Author of "AI Engineering", "Designing Machine Learning Systems"),

Simon Willison(Coauthor of Django), and

2 more.

CL4R1T4S by elder-plinius

1.3%

45k

Dataset of system prompts for major AI models + agents

System prompts from OpenAI, Google, Anthropic, xAI, Perplexity, etc.
Reveals hidden instructions affecting AI behavior and outputs.
Exposes personas, constraints, and ethical/political frames.

Created 1 year ago

Updated 3 weeks ago

Starred 1 day ago by

Travis Fischer(Founder of Agentic). Also starred by

Jane Manchun Wong(Security Researcher) and

Christopher Chedeau(Author of React Native, Prettier, CSS-in-JS; Cofounder of Excalidraw).

astryx by facebook

366.8%

A customizable, agent-ready design system for building consistent interfaces

Powers 13,000+ apps with over 90 accessible, themeable React components and templates.
Built with React and StyleX, featuring automatic spacing and a CSS variable cascade for theming.
Agent-ready tooling includes CLI and JSDoc annotations for AI-driven scaffolding and documentation.
Designed for both human developers and AI agents to build with unified tooling.

Created 5 months ago

Updated 1 day ago

Starred 1 day ago by

Xiaofan Luan(VP Engineering at Zilliz). Also starred by

Eric Zhu(Coauthor of AutoGen; Research Scientist at Microsoft Research),

Chip Huyen(Author of "AI Engineering", "Designing Machine Learning Systems"),

Alexander Wettig(Coauthor of SWE-bench, SWE-agent), and

5 more.

data-juicer by datajuicer

0.6%

Data-Juicer: Data processing system for foundation models

Library of 100+ data operators for cleaning, analyzing, and synthesizing data.
Supports multimodal data processing (text, image, video) and distributed processing.
Includes a sandbox environment for data-model co-development and recipe refinement.
Integrates with Ray for distributed data processing on cloud platforms.

Created 2 years ago

Updated 10 hours ago

Starred 1 day ago by

Christian Laforte(Distinguished Engineer at NVIDIA; Former CTO at Stability AI). Also starred by

Thomas Wolf(Cofounder of Hugging Face),

Alex Yu(Research Scientist at OpenAI; Cofounder of Luma AI),

Noah Snavely(Research Scientist at Google DeepMind; Professor at Cornell Tech), and

3 more.

ml-hypersim by apple

0.3%

Synthetic dataset for holistic indoor scene understanding research

Photorealistic images with per-pixel semantic instance segmentation.
Includes complete scene geometry, material, and lighting information.
Factors images into diffuse reflectance, illumination, and non-diffuse residual.
Leverages V-Ray for advanced rendering effects and high-quality 3D content.

Created 5 years ago

Updated 5 months ago

Starred 1 day ago by

Hahnbee Lee(Cofounder of Mintlify). Also starred by

Tomas Valenta(Cofounder of E2B),

Travis Fischer(Founder of Agentic),

Simon Willison(Coauthor of Django), and

3 more.

claude-code-action by anthropics

1.0%

GitHub Action for Claude to interact with code

Answers code questions, suggests improvements, implements fixes in PRs/issues.
Supports Anthropic API, Amazon Bedrock, and Google Vertex AI.
Uses GitHub APIs, file operations; allows custom tools via configuration.
Can fix bugs from screenshots using Claude's multimodal capabilities.

Created 1 year ago

Updated 2 days ago

Starred 2 days ago by

Jeff Hammerbacher(Cofounder of Cloudera). Also starred by

David Cournapeau(Author of scikit-learn) and

Joe Walnes(Head of Experimental Projects at Stripe).

local-llm by jamesob

N/A

916

Guide to building high-performance local LLM inference systems

Deploy SOTA LLMs and Whisper STT locally using Docker and vLLM.
Optimize multi-GPU setups with PCIe Gen4 switching for low-latency P2P communication.
Detailed hardware and BIOS/kernel tuning for maximizing VRAM and GPU throughput.
Cost-effective configurations from $2k (Qwen, Whisper) to $40k (near-Opus).

Created 3 days ago

Updated 2 days ago

...