Weekly Briefing

article sponsor image
Partner Content

Bring Your Data From People-Places-Things Together

November 22, 2023 | Justin Dupree, VP Customer Success, Mapped
4 min read
listen to article Listen to this article

Introduction
At Mapped, we take our commitment to simplifying data access seriously. To that end, we are introducing an innovative new feature called MappedGPT, empowering developers by providing them with an AI virtual agent capable of generating GraphQL queries from natural language.

MappedGPT is seamlessly integrated into our GraphQL Explorer, not only simplifying the query-building process but also offering immediate access to query results. We’ll dive into how we built MappedGPT, and explore the specific benefits it brings to Mapped developers.


Revolutionizing Data Access

Mapped leverages the power of machine learning to create an independent data layer, enabling property owners, facility operators, and solution providers to access real-time data from various sources. This data includes information from building systems, sensors, actuators, devices, and vendor APIs, automating the complex process of data discovery, extraction, and normalization.


Our unique approach to building data simplifies the task of extracting valuable insights from commercial and industrial spaces, enabling businesses to make data-driven decisions with ease. However, even with a simplified data integration process, developers still face the challenge of understanding what data is available, and how to construct effective GraphQL queries. This is where the new virtual agent comes into play.


Creating a Generative AI Virtual Agent

Our commitment to user-centric solutions led to the development of MappedGPT. We recognized developers using our platform needed a more efficient way to create GraphQL queries, eliminating the time-consuming process of manually crafting queries and reducing the potential for syntax errors. This new virtual agent is capable of automatically generating GraphQL queries based on natural language prompts.


Here’s how we created this game-changing feature:

  • Leveraging Machine Learning: MappedGPT leverages the power of LLMs and is trained on a vast dataset of GraphQL queries commonly used with our API. This training enabled the agent to understand the intent behind developers’ queries and learn how to generate corresponding GraphQL code automatically.

  • Seamless Integration with the Developer Console: We integrated MappedGPT into our GraphQL Explorer, ensuring developers can access it easily from the developer console, no need to load a new page or import an app. The integration is designed to be user-friendly, with a clean and intuitive interface that simplifies the process of interacting with the virtual agent.

Using MappedGPT

Load MappedGPT by clicking on the left side menu of the GraphQL Explorer, at which point you can “talk” to the API using natural language:

Describe the data you want to retrieve in plain English — the agent will translate these descriptions into GraphQL queries and show their output. For example, if the request prompt is written out like this:


“Retrieve temperature data for building ‘HQ’ on October 1st, 2023, including the timestamp.”


MappedGPT will generate the necessary GraphQL query in real-time:


Benefits for Mapped Developers

Some of the benefits MappedGPT brings to developers using the platform:

  • Faster Query Building: Significantly accelerates the GraphQL query-building process. Developers no longer need to spend time crafting queries from scratch. Instead, they can rely on MappedGPT to generate precise queries based on their descriptions.

  • Reduced Syntax Errors: Manual query creation often leads to syntax errors, which can be frustrating and time-consuming to debug. With MappedGPT, developers can avoid syntax issues altogether, as the generated queries are always syntactically correct.

  • Improved Productivity: By automating the query generation process, MappedGPT allows developers to focus on more critical tasks, such as analyzing data and building applications. This boost in productivity enables developers to deliver results faster.

  • Accessibility for Non-Technical Users: MappedGPT’s natural language interaction makes it accessible to a broader audience, including individuals with limited programming experience. Non-technical users can now easily request specific data without needing to learn the intricacies of GraphQL.

  • Immediate Query Results: One of the most significant advantages of MappedGPT is its ability to provide immediate access to query results through the GraphQL Explorer. Developers can test their queries and view the returned data within the same interface, eliminating the need to switch between tools and platforms.

  • Enhanced Collaboration: Collaboration between technical and non-technical team members is simplified with MappedGPT. Non-technical stakeholders can express their data requirements directly to the agent, fostering better communication within cross-functional teams.

  • Intuitive User Experience: Mapped’s commitment to user-centric design shines through in MappedGPT’s intuitive user experience. Developers can interact with the agent in a conversational manner, making it feel like a natural extension of their workflow.

  • Reduced Learning Curve: MappedGPT’s natural language interface reduces the learning curve associated with GraphQL for those new to the technology. Developers can get started quickly without needing to be experts in GraphQL syntax and can learn how to make more effective GraphQL queries by interacting with MappedGPT.

Conclusion

MappedGPT represents a significant improvement in data access and query building for developers. By leveraging advanced machine learning algorithms, seamless integration with the built in GraphQL Explorer, natural language interaction, and real-time query generation, this new tool streamlines the development process, reduces errors, and accelerates productivity.


The benefits extend beyond just developers, as it also empowers non-technical users to request specific data with ease. With immediate access to query results and an intuitive user experience, MappedGPT stands as a testament to our dedication towards simplifying complex data integration processes and making data-driven decisions more accessible to all.

Justin Dupree, VP Customer Success, Mapped
Justin Dupree has a diverse work experience spanning various roles and companies. Justin is currently the Vice President of Customer Success at Mapped since December 2020. Prior to that, they worked at Cisco from 2015 to 2020 as the Global Manager - Webex For Developers Support Engineering.

This Week’s Sponsor

Mapped started with the idea “what if every built space had an API?” Getting data out of buildings was a complex, time consuming and often manual process - weeks or even months of work. To solve the problem, we created a knowledge graph of people, places and things using machine learning to automate the process of data extraction and identification. We then built a simple, secure and unified API on top; now anything that generates data - devices, sensors, enterprise applications and more - is accessible quickly, easily and securely.