Frequently Asked Questions
Everything you need to know about Navsi AI
What is Navsi AI?
Navsi AI is a voice-native AI SDK that embeds inside B2B SaaS products. Users speak or type commands — Navsi autonomously executes complete workflows by navigating pages, filling forms, clicking buttons, and submitting data. Not a chatbot. Not a copilot. An autonomous executor that integrates in under 2 minutes with zero backend changes.
How is Navsi different from chatbots?
Chatbots answer questions and provide guidance, then hand work back to users. Navsi autonomously executes complete workflows — it navigates UI, fills forms, clicks buttons, and submits data. Users speak a command and the AI completes the entire task without requiring manual steps. The difference is execution vs. advisory.
How long does Navsi take to integrate?
Under 2 minutes. Install the SDK with `pnpm install @navsi.ai/sdk`, add environment variables (VITE_NAVSI_API_KEY and VITE_NAVSI_API_URL), create a wrapper component, and wrap your app routes. Zero backend changes required — Navsi works by understanding your application's frontend.
What SaaS verticals is Navsi built for?
Navsi targets B2B SaaS in 8 verticals: Legal (Clio, Smokeball, PracticePanther), Construction (Procore, BuildOps, Fieldwire), Healthcare (DrChrono, SimplePractice), Field Service (Jobber, ServiceTitan), Hospitality (Toast, 7shifts), PropTech (AppFolio, Buildium), FinTech (nCino, Blend), and HR Tech (Gusto, Rippling, Lattice).
Does Navsi require backend changes to my SaaS product?
No. Navsi requires zero backend changes. It embeds as a widget and works by understanding your application's frontend (DOM, accessibility tree, visual parsing). Your API endpoints, database schema, and backend logic remain completely unchanged.
Is Navsi voice-only or does it support text?
Navsi supports both voice and text input. It includes a built-in voice UI with microphone button and voice waveform visualization optimized for spoken commands, but users can also type commands if they prefer.
What are the two modes in Navsi?
Navsi operates in two modes: Ask Mode answers questions about your product and data (like a traditional chatbot), while Agent Mode autonomously executes complete workflows. Users can switch between modes or Navsi can intelligently determine which mode to use based on the command.
What languages does Navsi support?
Navsi currently supports English only. Additional language support is planned for future releases.
How do I get access to Navsi?
Navsi is currently in Beta with design partner slots available. You can book a demo at navsi.ai or join the waitlist. Demo calls are 20 minutes and focus on understanding your use case and product integration requirements.
How is Navsi different from Command AI or similar tools?
Tools like Command AI focus on search, shortcuts, and user guidance within your product. Navsi is an autonomous execution layer — it doesn't just help users find features, it completes entire workflows from voice or text commands. The core difference is voice-native autonomous execution vs. keyboard-driven navigation assistance.
Still have questions?
Book a 20-minute demo to see Navsi in action and get your questions answered.
Book a Demo