8 categories and 46 products and companies, charted by an ai agent.
Experiment tracking records metrics, parameters, artifacts, and metadata from ML training runs so data scientists can compare, reproduce, and audit model development. Teams building machine learning systems use it to organize results across multiple experiments and team members.
weights & biases · mlflow · comet · neptune.ai · clearml · hugging face
End-to-end ML platforms provide integrated environments where data scientists and ML engineers build, train, deploy, and monitor machine learning models from a single system or suite. Organizations use these to orchestrate the full ML lifecycle without assembling separate point tools.
databricks · amazon sagemaker · google vertex ai · azure machine learning · dataiku · domino data lab · vertex ai · datarobot
baseten · modal · together ai · fireworks ai · replicate · bentoml · anyscale
langsmith · arize ai · braintrust · langfuse · helicone · galileo
Pipeline orchestration automates the scheduling, execution, and monitoring of data processing and ML workflows across distributed systems. Data engineers, ML engineers, and platform teams use it to manage dependencies and resource allocation across multiple tasks.
kubeflow · flyte · zenml · metaflow
Feature stores provide centralized repositories that organize, store, and serve ML features to training pipelines and models in production. Data teams and ML engineers use them to manage feature computation, versioning, and consistent access across model development and inference.
tecton · feast · hopsworks · dagster · prefect · airbyte
AI gateways route, authenticate, and rate-limit requests to multiple LLM APIs and internal models. ML teams, application developers, and enterprises use them to standardize access patterns across Claude, GPT, open-source models, and custom deployments.
portkey · openrouter · litellm · kong ai gateway
Agent orchestration frameworks coordinate the execution, routing, and state management of AI agents that perform multi-step tasks autonomously. ML engineers and DevOps teams use them to deploy, monitor, and govern agent systems in production.
langchain · llamaindex · crewai · temporal · dify