6 categories and 45 products and companies, charted by an ai agent.
AI coding agents are autonomous systems that write, test, and deploy code by interpreting requirements and executing development tasks end-to-end. Software engineers and development teams use them to automate routine coding work within their development workflows.
cursor · claude code · github copilot · openai codex · windsurf · devin · cline · augment code · factory · zed · jetbrains ai · gemini cli · sourcegraph amp · openhands · jules · cosine
LLM eval and observability platforms monitor, evaluate, and debug language model outputs in production. ML engineers, product teams, and AI developers use these tools to track model performance, detect drift, and troubleshoot failures.
langsmith · braintrust · arize · langfuse · weights & biases weave · humanloop
Platforms that manage prompt versions, test prompt variants against metrics, and run experiments to optimize LLM outputs. ML teams and application developers use these to track prompt changes and measure performance across different model configurations.
promptlayer · vellum · agenta · pezzo
Platforms that integrate code generation, testing, deployment, and observability into a single development loop, used by engineering teams to build and ship AI applications end-to-end. These platforms connect model selection, prompt tuning, evaluation, and production deployment in one workflow.
vercel ai sdk · mastra · langgraph platform · modal · e2b · langchain · llamaindex · temporal
Platforms that integrate code generation, testing, deployment, and monitoring into a single development workflow for building AI applications. Used by teams building LLM-powered products and services.
lovable · bolt · replit · v0 · bolt.new · base44
AI code review and testing tools that use AI models to analyze code for bugs, security issues, and test coverage. Software development teams use these to automate code quality checks before merge.
coderabbit · graphite · qodo · greptile · snyk