Rasa logo

Rasa

Rasa is a platform for product teams building customer-facing AI agents, including chatbots and voice agents, that the company controls itself. The workflow starts with a prompt: describe the agent, then tune it in the user interface, in code, or with a coding agent, letting the model drive while the team adds order and conditions where they matter. Rasa evaluates the agent automatically, running it against simulated conversations in voice and text and showing what passed and what did not, so teams can run end-to-end checks, compare versions, and ship updates with confidence. After launch, Rasa reads conversations closely, tags and clusters what it finds, and surfaces recommendations on what to fix, such as where satisfaction drops as verification waits rise. Teams build together in the Studio workspace, where anyone can build, review, evaluate, and monitor agents visually while changes become code in a shared repository, and developers work in a structured, version-controlled codebase from their IDE with the CLI and SDKs. Coding agents can connect to the shared codebase through Rasa MCP Tools to automate builds, tests, and reviews. One agent can answer across a hotline, Slack, web chat, WhatsApp, email, and more, and the platform is built to deploy on the customer terms: their infrastructure, their model, and their data. Secure deployment, audit trails, and voice come standard, every decision can be traced for debugging, and the open, modular design lets teams swap pieces as priorities shift or run Rasa alongside existing agents. The pricing page lists the platform in depth, from dialogue understanding and management, a no-code flow builder, and language-agnostic natural language understanding to enterprise search, a contextual response rephraser, a custom actions server, and content and response management. Testing tools include end-to-end testing, a try-your-assistant panel, conversation views, rollback, an LLM fine-tuning recipe, and multi-LLM management. Deployment runs on Kubernetes with Helm support, Redis concurrency, a Kafka data pipeline, and OpenTelemetry observability, while governance covers single sign-on, role-based access control, PII data management, Vault-powered secrets management, and dependency vulnerability protection. Pricing has two tiers and no published figures for the paid one. The Free Developer Edition can be used locally or in production for one bot per company, within monthly conversation limits, with community support through the forum. Enterprise adds full platform access, premium support with a customer success manager and engineer, success planning, and best practice guidance, and is sold through sales. Customer stories on the site span healthcare, finance, telecom, retail, and insurance, which suits regulated and enterprise teams that want a self-hosted, developer-owned alternative to a managed black box agent.

Automation Customer Support

Similar listings in category