No credit card required
At RAW, our belief is that the future of business intelligence lies in the seamless integration of conversational interfaces with data-driven insights, enabling stakeholders to interact with and query business data in a more natural and intuitive way.
In the past, we discussed our approach to the integration of LLMs with operational data, which informed our view on the rise of chat and the evolving role of websites. We believe that we are about to see significant changes to how we interact with and extract value from data in a world where chat interfaces are becoming the primary mode of digital interaction.
But how should companies prepare? What should they do to become the first movers that reap the benefits and build unique differentiation? This blog gives a practical example by showing how we do it at RAW Labs—as we "eat our own (dog) food" so to speak.
Welcome then, to a short introduction to the "RAW Labs API Operating System" (API OS).
The API Operating System: Harnessing Real-Time Data and LLMs
APIs, or Application Programming Interfaces, are the backbone of digital communication, enabling different software applications to interact with each other seamlessly. At RAW Labs, we have developed a comprehensive system of APIs that cover all essential parts of the business, allowing us to observe and ask, in real time, what is happening.
At the heart of the API Operating System is the integration of real-time operational data with LLMs. With it, we leverage the power of LLMs to analyze, interpret, and respond to data queries in real-time, providing insights that are not just timely but also contextually rich and nuanced.
The RAW Labs API OS
At RAW Labs, the architecture of the API OS is modular, with clusters of APIs focused on distinct business domains. For instance, within the sales domain, the API OS might integrate a suite of APIs that query CRM systems, delve into marketing platforms, or even query ongoing work by the dev team, creating a comprehensive view of the sales ecosystem. These APIs are engineered to tap into a myriad of data sources, both structured and unstructured, delivering insights with speed and relevance.
Our API OS extends across all critical business functions support; here is a brief view of the API modules we currently run internally:
- Company: This module of APIs offers a panoramic view of RAW Labs, from our foundational ethos to the strategic milestones shaping our journey.
- HR: This module provides a real-time snapshot of our workforce structure and dynamics, covering everything from "who is away from work today" to sensitive employee information.
- Product: This module delivers deep insights into our product's lifecycle, architecture, and roadmap. For this it interacts both with documentation we produced as well as operational systems used in development - e.g. JIRA, Confluence, GitHub, etc.
- User Engagement: As a SaaS platform, it is essential to track the user activity and various KPIs in real-time, which this module does, by providing a wide-range of metrics that are specific to our business model.
- Marketing: Provides an overview of key marketing activities, including a real-time view of outbound activities and tracks core KPIS.
- Sales: Offers a real-time pulse on our sales performance, including data from CRM, marketing sources, etc.
- R&D: Tracks real-time development progress, to help us ensure strategic alignment and operational efficiency.
- Support: Provides a real-time view on support issues, allowing us to complete a 360 view of the client's journey.
- Operations: Tracks status of product, services and costs, in real-time.
- Finance: Tracks key financial metrics, and even details aspects such as our SaaS or cloud costs.
The Chat Interface: A Gateway to Business Intelligence
Embracing the shift towards chat interfaces, our API Operating System is then used from a chat-based interface such as ChatGPT, allowing our employees to interact with our company's information in a conversational manner. This aligns with our vision, as outlined in our exploration of chat's rise, recognizing that the future of digital interaction is increasingly conversational and user-centric.
Let's now give a real example; here's a question asked to our API OS via ChatGPT:
The Power of Real-Time Data Access
By harnessing the RAW platform's capabilities, these APIs can retrieve and synthesize data from diverse sources on-the-fly, offering a dynamic and up-to-date picture of the business. The data is then fully contextualized with metadata by RAW, so that the LLM understands and can use it; this includes describing the available data, its sources, necessary input parameters, and how to process the outputs effectively. This clarity is crucial, enabling the LLMs to leverage the API OS for generating nuanced, context-aware insights.
Hosting, Security, and Maintenance
The RAW platform allows you to create these potent APIs with ease, but also hosts and secures them, offering a turnkey solution for businesses looking to harness the power of real-time data. This is what we use internally, where the individual modules have different permissions and access control, so that different roles in the company can be exposed to the information they need. We provide fine-grained control over access, as well as contextualized answers, so that C-level sees responses at the right-level of detail and abstraction for example.
Maintenance is also extremely important because, as a startup, our operations can evolve rapidly. Here we rely on many RAW features; for instance, the platform automatically generates and maintains a data catalog or OpenAPI specification, which is essential for the LLMs to interact with the APIs efficiently. This automated upkeep ensures that the APIs remain robust, secure, and up-to-date, allowing businesses to focus on leveraging the insights they provide rather than worrying about their maintenance.
A New Era of Business Intelligence?
The API OS we use at RAW Labs is an example of the upcoming era of business intelligence, where real-time data access, modular composable designs, and seamless integration with LLMs converge to create a powerful tool for decision-making. This approach not only enhances the agility and responsiveness of businesses but also unlocks new potential for innovation and strategic insights.
Next Steps: Let's Build it?
Upcoming posts will describe in detail how we built our API OS, step-by-step. But in the meantime, if you want to learn more, or even try it out yourself, reach out to us, or book a demo, as we are happy and excited to assist you in your own API OS journey!
