Why Taros is modular by design
Simple dashboard setup, multiple AI models, reusable data sources, and an API when you need it. Here's how Taros stays flexible as you grow.
Most AI platforms force you to choose: either click-and-drag simplicity that hits a wall when you need customization, or developer-first complexity that requires a CS degree to get started.
We think that's a false choice.
No-code that actually works
Want to get started without writing a single line of code? Perfect.
Simple setup. Upload your documents, crawl your website, customize your bot's personality – all through our dashboard.
Website widget. Deploy a fully customizable chat widget to your site with a single script tag. Configure colors, position, animations, welcome messages, and more – all visually.
Multiple training methods. Crawl your website, upload documents (PDFs, Word, Excel), or manually add Q&A pairs. Mix and match to build the knowledge base your assistant needs.
Most customers get their first AI assistant running in under 10 minutes using just the dashboard.
Pick your AI model
Taros isn't locked to a single AI provider. Choose the model that fits your needs:
Mistral is available on all plans and works great for most use cases. If you need more power, OpenAI and Anthropic models are available on higher-tier plans.
Switch models anytime without rebuilding your assistant. Your training data, settings, and integrations stay the same.
Built for teams
Multi-user organizations. Invite team members with role-based access. Your marketing team can manage content while developers handle integrations.
Analytics that matter. Track conversation topics, unanswered questions, purchase signals, and resolution rates. See what's working and where to improve.
Continuous improvement. Review chat history, identify gaps in your knowledge base, and add manual training to fill them – all from the dashboard.
When you need more control
The dashboard handles most use cases, but we also have a REST API with 100+ endpoints for when you need programmatic control.
API access. Manage assistants, data sources, conversations, and more through our API. Create API keys directly from the dashboard.
Webhooks. Get notified when events happen in your assistant – new conversations, form submissions, escalations. Available on Pro.
Interaction flows. Build multi-step forms with conditional logic to capture leads and route conversations.
Modular by design
Taros is built around the idea that different parts of your setup should be independent and reusable:
Data sources are separate from assistants. Create a knowledge base once and assign it to multiple assistants. Update content in one place, and all connected assistants stay current.
Flexible deployment. Deploy as a website widget today, add Shopify integration when you need it. Discord and email integrations are coming soon.
Search tuning. Control how your assistant finds answers with relevance scoring, priority weighting, and configurable result limits.
The philosophy
You shouldn't need a computer science degree to set up customer support AI. But you also shouldn't be stuck with a rigid system that can't grow with your business.
Start simple with the dashboard. Use the API when you need automation. Add new channels as they become available. Each piece works independently, and they all work together.
That's the Taros way.