Language
HomeLanguages
Sections
Neural networks generate wobbly pixel art with inconsistent grid sizes. The perfectPixel library fixes this using FFT and classical computer vision—no heavy diffusion required.
Lightweight face tracking library optimized for CPU, delivering 30–60 fps even on older hardware without heavy neural network frameworks.
A lightweight local Gmail cleanup tool with a web interface that works entirely on your machine via the official Gmail API—no subscriptions or third-party servers required.
NomadNet creates a fully autonomous mesh network without internet, IP addresses, or providers using LoRa, radio modems, or even serial interfaces.
You're the OS! turns OS scheduling concepts into a hardcore arcade game where you manually manage CPU, memory, and I/O interrupts to keep angry users from hitting Reset.
Stop wasting evenings on FastAPI boilerplate. Meet s3rius's interactive project generator that builds a production-ready backend skeleton in minutes with your chosen stack.
Turn a local Ollama language model into an autonomous research agent that digs deep into topics using DuckDuckGo searches and iterative analysis.
Hyper-Extract converts messy documents into structured knowledge graphs and typed data using ready-made templates, eliminating the need for complex prompt engineering.
Fed up with fitness app subscriptions? Here's how I deployed the open-source wger workout and nutrition manager on my own server.
Tired of wrestling with Postfix configs and spam filters? Meet Mailu — a Docker-based mail server that deploys in minutes with all the essentials built in.
Omnigent is an open-source Python meta-framework that orchestrates multiple AI coding agents under one roof with unified interfaces and security policies.
Centaur transforms local AI agents into shared team assistants living in Slack. Built by Paradigm with a security-first approach using isolated sandboxes.
A practical guide to combining Obsidian and Claude Code for building a personal knowledge base that stays organized without vendor lock-in.
txtai is an all-in-one Python library combining vector database, hybrid search with SQL, graph analytics, and LLM orchestration for RAG pipelines.
SD.Next is a fork of Automatic1111's WebUI that adds cross-platform support, efficient resource management, and built-in tools for image and video generation.
A CLI extension for Claude Code that automates SEO audits, generates actionable roadmaps, and validates technical issues directly in the terminal.
Harvey LAB is an open framework and dataset for evaluating AI agents on real legal tasks, featuring over 16,000 scenarios across 20+ practice areas.
Building AI agents with local LLMs? SIE (Superlinked Inference Engine) consolidates multiple inference servers into one, cutting GPU costs and simplifying your container setup.
DeepMind trained neural networks to predict atmospheric processes in seconds, releasing WeatherNext 2, GraphCast, and GenCast with specialized cyclone models.
Vector databases and standard RAG can't answer auditor questions about AI decisions. Semantica adds deterministic graphs, rules, and audit trails.
SkyRL is an open-source RL fine-tuning framework from Berkeley that lets you train LLM agents on your own GPUs with async weight updates and built-in environment support.
When popular AI editors raised prices again, developer Mark Pfaffenberger built Code Puppy — a console AI agent that replaces closed IDEs with API key rotation and deep customization.
A deep dive into ComfyUI's workflow_templates repository—how it transforms JSON graphs into self-contained template bundles with model bindings and automated localization.
Mypy catches type errors before code runs, protecting Python projects from runtime crashes. Here's why gradual typing is worth adopting today.