Datalang

DataLang has developed a unified platform that enables custom chatbot applications to integrate data from multiple sources, including SQL databases, files, Notion, and Google Sheets. This comprehensive solution abstracts the complexities of data integration while maintaining rigorous security standards, making it accessible for developers of all skill levels. The platform's unique approach to secure data exposure and its flexible deployment options make it an attractive solution for building conversational applications that require access to diverse data sources.

Supported Data Sources

DataLang's platform supports an extensive array of data sources, including SQL databases, files, Notion, Google Sheets, and several additional options. For database integration, the platform works with PostgreSQL, Content (website), and various DSN options using ODBC. It also supports future developments for Snowflake, MySQL, MariaDB, Oracle, Amazon Redshift, and SQLite.

The company's approach to data integration centers on users with read-only access to the desired tables, ensuring secure data exposure. Data credentials are encrypted, and decryption occurs only during syncing processes. The platform provides a standardized API endpoint for accessing integrated data: datalang.io/api/data-sources/{source-slug}/views/{view-slug}?sync=false.

The four-step development workflow focuses on straightforward implementation, as detailed in the documentation. This process enables users to integrate data from various sources into their chatbot applications efficiently.

Data Exposure and Security

Data exposure follows a secure model where credentials remain encrypted unless actively syncing. During development and operation, the platform maintains a separation between data at rest and in transit to prevent unauthorized access. The standardized API endpoint (datalang.io/api/data-sources/{source-slug}/views/{view-slug}?sync=false) facilitates secure data retrieval while abstracting the complexities of credential management from the end-user.

The company's approach balances security with functionality by requiring users to have read-only access to the specific tables they wish to expose from their data sources. This targeted approach ensures that sensitive data remains protected while enabling relevant information to be integrated into chatbot workflows.

Chatbot Development Workflow

The development workflow consists of four straightforward steps. First, users set up their data sources, which can include SQL databases, files, or website content. Integration with Notion, Google Sheets, and other data sources follows this same read-only access principle, with credentials encrypted during transit and decryption reserved for syncing processes.

Step two involves adding data views to expose specific data elements from the selected sources. These views act as secure data layers, allowing developers to fine-tune which information is accessible to their chatbots while keeping the underlying data protected.

The third step focuses on configuring chat functionality. This involves training the platform's GPT-3 engine with the selected data sources to create a conversational foundation for the chatbot. The platform abstracts the technical complexities of data integration, allowing users to focus on refining their chatbot's conversational capabilities.

The final step enables users to share their chatbots through one of four options: a public URL, embedding the chatbot widget on a website, publishing to the GPT Store, or integrating with their own API endpoints for custom question submission. These sharing methods accommodate both individual developers and enterprise-level deployments, with pricing tiers scaling from free to business-level support to meet various organizational needs.

Integration and Sharing Options

Data management and sharing in the DataLang platform emphasize both flexibility and security. Users can deploy their chatbots through multiple channels, with options that range from public visibility to integrated backend access.

The public URL method provides immediate access to chatbots, making them instantly available to anyone with the link. This approach is useful for sharing basic chatbot capabilities without the need for additional integration or embedding.

For website integration, DataLang offers a chatbot widget that can be embedded directly into web pages. This option allows businesses to host the chatbot interface on their own website, providing a seamless user experience while maintaining control over their platform.

Organizations can also publish their chatbots to the GPT Store, which serves as a repository for third-party chatbot applications. This method enables wider distribution and discovery of the chatbot while giving developers access to additional user base.

The platform's API integration option allows for more advanced deployment scenarios. Developers can configure the chatbot to accept questions through their own custom systems, providing flexibility for complex or specialized applications.

Pricing for these deployment options scales with the needs of the organization. The free tier supports basic usage for individuals and small teams, while the business plan provides advanced features and priority support for enterprise-level implementations.

Pricing and Plan Details

Pricing options scale to meet the needs of individuals and organizations of all sizes. The free tier supports basic usage with one user account, ability to integrate one data source, and includes 100 monthly credits. This level provides access to the chatbot widget feature and no customer support.

The Basic plan raises the user limit to two individuals while expanding data source capabilities to ten. It offers 1,000 credits per month and continues to include the chatbot widget. Moving to the Pro plan increases support capacity to six users and data source limits to fifty. Monthly credits increase to 3,000 while providing basic support services.

The Business tier represents the highest level of access, supporting twelve users and one thousand data sources. This plan includes the most extensive credit allocation at 20,000 monthly credits and provides priority support services. All pricing tiers allow for sharing chatbots through the public URL, embedded widget, GPT Store publication, and API integration methods described in the documentation.