May 2024 Summaries

Unified.to has announced significant product updates for May 2024, featuring over 190 integrations across 15 API categories, including CRM, E-commerce, and Accounting, making it the fastest unified API platform for adding product integrations. Key updates include the launch of the Unified Messaging API, allowing access to customer communications from platforms like Slack and Microsoft Teams through a single API, and the Unified Knowledge Management System (KMS) API, which integrates with tools like Notion and Atlassian Confluence for enhanced content access and predictive analysis. Enhancements to the Unified Accounting API now include additional fintech integrations and support for real-time transaction entry, while the Unified Authentication API has expanded to offer SSO capabilities with over 60 identity providers without third-party data storage. A new UI update introduces an Actions feature for efficient integration testing and deployment and a Category dropdown for easier browsing of API options, emphasizing Unified.to's commitment to rapid integration deployment and user-centric improvements.

Unified.to's GenAI API offers a streamlined way to interact with various large language models (LLMs) from providers like OpenAI, Anthropic, and others through a unified interface. The guide explains how to set up and use the GenAI API by integrating it with platforms such as OpenAI's GPT and Anthropic's Claude, allowing users to send prompts and receive responses from these models. The setup process involves creating connections, acquiring API tokens, and configuring environment variables. Users can experiment with different model parameters, such as temperature, to influence the creativity of the AI's responses. The API supports interacting with multiple LLMs, enabling flexibility in selecting models based on complexity and cost. Unified.to's API is particularly useful for SaaS companies looking to integrate generative AI into their services, and it is available on all Unified.to plans, including a free 30-day trial.

Unified.to has introduced a new category within its API for knowledge management systems (KMS), allowing applications to integrate and leverage content from platforms like Atlassian Confluence, Coda, Guru, and Notion through a single API. This unified API abstracts the variations in content organization across different KMS platforms, simplifying access and management by categorizing information into 'Space' and 'Page' objects. The API enhances the accessibility and utility of institutional knowledge, which often becomes underutilized due to organizational changes and everyday business operations. By pairing the KMS API with large language models (LLMs), users can automate tasks such as generating meeting summaries, creating documentation, and transforming natural language into structured data formats. This integration facilitates real-time Q&A, knowledge discovery, and the conversion of textual notes into more processable data forms, opening up new possibilities for data analysis and insight generation. The KMS API is available across all Unified.to plans, including the free one, with easy integration into LLMs and over 190 SaaS platforms, aimed at simplifying knowledge extraction and enhancing business intelligence.

Unified.to's Generative AI API offers a streamlined approach for interacting with various large language models (LLMs) from providers such as OpenAI and Anthropic Claude. The guide walks users through setting up the necessary accounts and tools, such as Python and Jupyter Notebook, to activate and connect these integrations via Unified.to's platform. By employing the Unified.to API, users can send prompts to these LLMs and receive responses, adjusting parameters like temperature for varied output. The API also supports querying available models, allowing users to choose based on complexity and cost. Additionally, the guide highlights the potential of integrating this generative AI capability with other SaaS applications like ATS, KMS, messaging, and storage systems to analyze and process data, emphasizing the API's utility in real-world applications across different sectors. Unified.to's GenAI API is accessible on all its workspaces, with a free 30-day unlimited-use Tester plan to facilitate easy onboarding and experimentation.

The Unified KMS API provides a streamlined solution for accessing and managing knowledge content across various platforms by offering a standardized interface that normalizes content structures and data models. This API serves as a cohesive knowledge layer, allowing applications to interact with documentation, help center content, and knowledge bases through a consistent object model that includes spaces, pages, and comments. It supports real-time access and updates by routing requests directly to source platforms without storing customer data, and it utilizes webhooks for content changes. The API's integration capabilities span platforms like Atlassian Confluence, ClickUp, Notion, and Salesforce, and it is particularly beneficial for internal tools, support platforms, search systems, and AI applications, enabling them to build unified search functions, aggregate help center content, and use documentation for AI-driven tasks. Designed for developers, it simplifies the retrieval and management of documentation by standardizing content access, structuring, and updates across multiple providers, and it is available on all Unified plans with options for trials and demos.

Setting up OAuth 2 credentials for Greenhouse's Candidate Ingestion API involves several steps that are crucial for enabling Unified.to to authorize access to users' Greenhouse accounts. Initially, it is important to differentiate between Greenhouse's Candidate Ingestion API, which uses OAuth 2, and the Harvest API, which does not. To proceed, you must provide Greenhouse with essential application details, including the application name, URL, callback URL, and a logo image. Upon submission, Greenhouse will issue a consumer key and a consumer secret, the latter being encrypted for security purposes. Once these credentials are obtained, you can activate the integration within Unified.to by entering the consumer key and secret into the appropriate fields in the Active Integrations page, allowing OAuth 2 authorization to be fully operational.

Setting up OAuth 2 credentials for Greenhouse's APIs involves choosing between two integrations on Unified.to: the Harvest API and the Candidate Ingestion API. The Harvest API, which uses OAuth2 code-flow, is suitable for most use cases and requires companies to submit an integration request form, reviewed by Greenhouse monthly, unless they are an official partner. Once approved, companies receive client credentials and must configure their integration by providing a redirect URI, which Unified.to uses to manage authorization processes. The Candidate Ingestion API uses OAuth 2 client-credentials-flow, necessitating the submission of application details to Greenhouse, including application name, URL, callback URL, and logo image. Upon approval, Greenhouse provides a consumer key and secret, which are used to activate the integration in Unified.to by entering these credentials on the Active Integrations page and clicking the Activate button.

To integrate Discord with Unified.to, users must first register their application in Discord's Developer Portal by creating a new application and agreeing to the terms of service. Once registered, they can obtain the OAuth 2 credentials, including the Client ID and Client Secret, necessary for authenticating the app with Discord's APIs. Users must also retrieve a Bot Token from the Bot section, ensuring they store it securely as it cannot be viewed again. Additionally, specific bot permissions must be enabled depending on the API endpoints used, such as Server Members Intent and Message Content Intent. Finally, users can activate the Discord integration within Unified.to by entering the OAuth 2 credentials and Bot Token, allowing them to leverage the Unified API for enhanced Discord experiences.

Unified.to has introduced a new Messaging API that consolidates access to chats and emails from popular platforms like Discord, Google Mail, Microsoft Outlook, Microsoft Teams, and Slack into a single interface. This API abstracts the differences between various messaging providers and categorizes both emails and chat messages as a unified data type, enabling seamless retrieval and analysis of messaging data. By leveraging large language models (LLMs), the Messaging API can summarize corporate discussions, identify and generate tickets from informal requests, automate responses, and derive customer insights through sentiment analysis and trend identification. Available across all Unified.to plans, including the free tier, this API allows users to integrate with over 170 SaaS applications and provides real-time data for building insightful applications, with a special offer of a 30-day unlimited-use Tester plan for new users.

Setting up and configuring a public Notion application involves registering the integration, understanding the difference between internal and public integrations, and using the Unified API to query pages and databases. Internal integrations are confined to a single workspace, while public integrations are accessible across multiple workspaces and use the OAuth 2.0 protocol for access. Registering a public integration on Notion's platform requires creating the integration, obtaining OAuth credentials, and selecting necessary permissions. To query databases, the Unified KMS endpoint is used, where database entries are treated as pages and can be sorted and filtered with various parameters. Notion does not offer programmatic webhook creation, so users must manually create them via the Notion integrations page, specifying a public Webhook URL and selecting the desired events. Once the webhook is created, Notion sends a verification token that must be validated to finalize the subscription process, allowing real-time event reception through Unified.to.

The Unified Messaging API offers a streamlined solution for accessing and sending email and chat data across various platforms by providing a single interface with a normalized data model and real-time access. It integrates with popular services like Gmail, Slack, Microsoft Teams, and more, and standardizes messages by using consistent schemas for emails and chat messages, threads, participants, and attachments, allowing seamless communication across different providers. By acting as a unified communication layer, it facilitates the creation of unified inboxes, customer support systems, and internal communication analysis tools, as well as AI systems that analyze conversations for insights and trends. The API supports real-time data access without the need for polling or storing message data, using webhooks to deliver updates, making it particularly beneficial for customer support platforms, CRM tools, and automation systems that require immediate interaction with communication platforms.

Unified.to has unveiled its generative AI (GenAI) API, which integrates multiple large language models (LLMs) like Anthropic Claude, Google Gemini, and OpenAI GPT into a single unified object model, simplifying application integration by eliminating the need for separate coding for each LLM. This GenAI API, when used alongside Unified.to's real-time API that connects with over 190 SaaS applications, enhances automation, data analysis, and personalized interactions by leveraging real-time structured data. The API supports Retrieval-Augmented Generation (RAG) to improve LLM output by supplementing it with current external data, increasing accuracy and allowing developers to tailor results according to specific needs. The integration offers various business solutions, such as predictive analytics and personalized messaging, across domains like recruitment, customer relationship management, and marketing, and is available on all Unified.to plans, including a free 30-day trial.

The Unified Generative AI API offers a streamlined solution for integrating with large language models (LLMs) by providing a single interface to send prompts and receive responses across multiple providers, such as OpenAI, Google, and Anthropic. This API standardizes interactions with LLMs, ensuring consistent request structures and response formats, which simplifies switching between providers without altering integration logic. It enables real-time model execution by routing requests directly to providers and supports retrieval-augmented generation (RAG) by allowing models to access live data from external systems at request time. This functionality allows AI systems not only to generate responses but also to retrieve current data and trigger actions in external systems, effectively operating on live context. As a unified solution, it aids AI product teams, platforms, and applications in building systems that require dynamic model routing, integration with operational data, and execution of tasks across APIs.

A unified API consolidates multiple APIs from various platforms into a single interface, simplifying integration for developers by reducing complexity and development time, while also enhancing functionality and expanding an application's market reach. This approach standardizes interactions across different services, allowing developers to use consistent methods for data manipulation and reducing the need for multiple tools and systems. Features such as unified endpoints, data models, authorization, scopes, and webhooks are crucial in a unified API, providing a seamless and secure experience. Security remains a priority, with best practices including data encryption and secure storage of credentials. The choice of a unified API should consider compatibility with specific use cases, breadth and depth of supported vendors and data fields, as well as an appropriate pricing model that aligns with the application's growth and business model. Unified.to exemplifies these principles, offering real-time data access, extensive vendor support, and robust security features to cater to diverse use cases.