Teaflask
Teaflask provides an AI agent platform that companies embed directly into their own software products, giving users an in-product assistant that understands the product and executes tasks on their behalf. The agent is grown from the host company's own codebase, enabling it to operate within the product's rules and workflows, such as filing expenses from receipts or managing travel bookings. The platform is free to use, with customers only paying for the underlying AI model at cost with no markup.
- AI Agents
- Artificial Intelligence
- Enterprise Software
- Software Only
Funding
Raised to date
$500KRaised to date based on public sources. This may differ from the amount the company actually raised and is based only on what is publicly available on the internet.
Founders
Founder details are not available yet.
Product
Problem
Users waste time navigating menus, configuring settings, and dealing with complex software interfaces to accomplish what they already know they want to do. This interaction overhead, which Teaflask describes as a "tax" on user intent, leads to friction, support tickets, and lost productivity as users translate their decisions into the specific steps required by the software.
Solution
Teaflask provides an AI agent platform that companies embed directly into their own software products, giving users an in-product assistant that understands the product and carries out tasks on their behalf. The agent is grown from the host company's own codebase, so it understands how the product works and can operate within its rules and workflows. Users express intent in natural language, and the assistant handles the execution—such as filing expenses from receipts, managing travel bookings, or submitting reimbursements—while providing a clear receipt of every action it takes. The platform includes pre-loaded skills for common back-office operations, and the assistant reads page context to understand what the user is trying to do, with sensitive data redacted and approval required for consequential actions.
Target Audience
Primary customers are SaaS companies that want to embed an AI assistant into their products to automate user workflows, particularly those with back-office functions like expense management, travel, and accounting.
Features
- Embedded assistant built from the host product's codebase, enabling domain-specific understanding and native workflow integration
- Pre-loaded skills for expense filing, receipt processing, travel management, and other back-office operations
- Context-aware page reading that captures visible content while redacting password fields, card numbers, and one-time codes
- Approval-based action execution where consequential operations require user confirmation under host-configured rules
- No-cookie, cookieless embedded assistant with short-lived in-memory tokens and namespaced local storage entries
- Optional ChatGPT integration allowing users to connect their own plan to power assistant conversations
- Model-agnostic architecture supporting multiple Claude models billed at cost with no platform markup