Build and Ship

IT Agent is an IT management platform offering remote access, automated patching, and task automation. It is designed for IT admins, MSPs, and teams seeking fast and configurable remote access. The platform provides endpoint monitoring and tools to manage IT operations efficiently.

Qubrid AI is an open full-stack AI platform designed for fast and scalable inference of AI models. It unifies GPU cloud, model serving, autoscaling, and production inference into a single platform. Qubrid allows users to deploy, optimize, and scale inference on NVIDIA GPUs. It supports Kubernetes orchestration and offers API and browser access.

StatusDude is a monitoring tool for websites, APIs, Kubernetes clusters, and internal services. It automatically discovers and monitors infrastructure, including resources behind firewalls. The tool offers multi-region verification to reduce false positives. It features a free tier for up to 7 monitors.

PyTriton is a framework that simplifies the deployment of NVIDIA's Triton Inference Server in Python environments. It offers a Flask/FastAPI-like interface for serving machine learning models directly from Python. PyTriton streamlines model deployment and optimization for performance.

Logparser is a machine learning toolkit for automated log parsing. It extracts event templates from unstructured logs and converts raw log messages into structured events. This tool is designed to streamline log analytics by automating the process of message template extraction and log key extraction.

Loglizer is a machine learning toolkit designed for log-based anomaly detection. It helps automate the process of identifying unusual patterns and potential issues within system logs. The tool supports automated anomaly detection and intelligent fault diagnosis, making it useful for maintaining software systems.

LLM-Viewer is a tool for visualizing Language and Learning Models (LLMs). It analyzes performance on different hardware platforms. The tool enables network-wise analysis, considering factors like peak memory consumption and total inference time cost. It is available as a web browser or command line interface (CLI) tool.

llm-server-docs provides end-to-end documentation for setting up a local and private LLM server on Debian. It includes features like chat, web search, RAG, model management, image generation, and TTS. The documentation covers configuring SSH, firewall, and secure remote access.

MLRun is an open-source MLOps platform designed for building and managing continuous ML applications. It integrates into development and CI/CD environments. MLRun automates the delivery of production data, ML pipelines, and online applications across their lifecycle.