AI Topic: Infrastructure

Koyeb is a developer-first serverless cloud platform for deploying containers, APIs, and ML inference across GPUs and global edge locations with automatic scaling and pay-per-use pricing.

A pure JavaScript port of Google's JAX ML framework that compiles and runs neural networks directly in the browser via auto-generated WebGPU kernels—with autodiff, JIT, and vectorization built in.

A public directory and index of Model Context Protocol (MCP) servers that lets developers discover, browse, and integrate data and tool connectors for AI agents and applications.

AI observability platform for building, evaluating, monitoring, and shipping quality AI products.

AI cloud platform for training, running, and serving models with pay-per-token APIs, managed GPU infrastructure, and collaborative cloud workspaces.

Tinker is an API for efficient LoRA fine-tuning of large language models—you write simple Python scripts with your data and training logic, and Tinker handles distributed GPU training.

Open-source-first AI company providing state-of-the-art LLMs, APIs, agent orchestration, fine-tuning tools, and deployable private inference solutions for enterprise and developer use.

Open-source AI desktop and web app that runs local and cloud models, provides assistants, connectors, and a local OpenAI-compatible API for self-hosted workflows.

Snyk is an AI-powered application security platform that finds, prioritizes, and helps fix vulnerabilities across code, open source dependencies, containers, infrastructure-as-code, and APIs.

A web-hosted ML model registry from Cloudsmith for storing, versioning, and managing machine learning models and related artifacts with API access.