Inspiration

As a Google Maps Premium Partner in Latin America, we've had a front-row seat to the digital transformation of countless businesses. We've seen logistics companies striving for efficiency, financial institutions battling fraud, and retailers aiming to create seamless customer experiences. In every case, the immense power of Google Maps Platform was the key.

However, we consistently saw the same barrier: a "complexity gap". Brilliant business ideas were getting bogged down by the technical challenge of integrating and orchestrating multiple, powerful APIs. Development cycles were long, costs were unpredictable, and the true potential of location intelligence remained just out of reach for many.

Our inspiration was born from this experience. We asked ourselves: What if we could build the tool we wish our clients had? What if we could democratize access to this power? We were inspired to create a platform that would eliminate this friction, acting as a universal translator between complex business needs and the powerful Google Maps ecosystem.

What it does

GeoServi Core is an intelligent API gateway and orchestration platform. It functions as a smart abstraction layer, providing a single, intuitive endpoint for developers to access the full spectrum of Google Maps Platform's capabilities.

Instead of forcing developers to learn and manage multiple complex APIs, GeoServi Core provides a set of standardized, business-oriented "tools" that can be called upon. For example, a developer doesn't need to figure out the intricacies of the Distance Matrix API; they simply ask GeoServi Core to use the get_distance_matrix tool.

Our platform:

  • Orchestrates API Calls: Intelligently selects the correct Google API (or combination of APIs) to fulfill a request.
  • Optimizes Consumption: Reduces redundant calls and is designed to use the most cost-effective SKU for the job.
  • Standardizes Responses: Returns clean, predictable, and ready-to-use JSON, drastically simplifying frontend and backend development.
  • Acts as a Tooling Protocol for AI: It's designed from the ground up to be consumed by AI Agents and LLMs, providing the "grounding" they need to interact with the real world reliably.

How we built it

We built GeoServi Core on a modern, scalable, and secure architecture, designed for enterprise-grade performance.

  • Tech Stack: The core is built in Python using the high-performance FastAPI/Starlette framework, which is ideal for asynchronous, high-throughput API services.
  • Architecture: The system is composed of two main components:
    • A Server MCP that directly interfaces with the official Google Maps Python libraries, exposing the standardized "tools".
    • A Client MCP that acts as the intelligent agent, receiving queries, interpreting intent, and calling the appropriate tools on the server.
  • Stateless Design: The entire platform is stateless, meaning each API request is independent. This is crucial for security, reliability, and horizontal scalability.
  • Deployment: The service is containerized and designed for cloud-native environments, running seamlessly on Google Cloud Run, which allows for automatic scaling from zero to thousands of requests.

Challenges we ran into

One of our primary challenges was designing the intelligent tool selection logic within the Client MCP. Initially, it was based on simple keyword matching, but we quickly realized this was too rigid. The real breakthrough came when we evolved the logic to understand the semantic intent of a query, allowing for much more flexible and accurate tool orchestration.

Another significant challenge was defining the standardized JSON response schema. We had to create a structure that was rich enough to convey all the necessary data from different APIs (like routes, places, and validation results) but also simple and consistent enough that a developer could consume it without needing to write custom parsers for each tool. This required several iterations to achieve the right balance between flexibility and predictability.

Accomplishments that we're proud of

We are incredibly proud of creating a platform that truly lowers the barrier to entry for sophisticated geospatial development. Our greatest accomplishment is seeing the "before and after" effect: a task that previously required a developer to write over 100 lines of complex code can now be accomplished with a single, simple API call to GeoServi Core.

Before: Complex, multi-API integration

directions_client = DirectionsService()
places_client = PlacesService()
# ... extensive setup and logic ...

After: One simple, intuitive call

response = GeoServiCore.query("Find the fastest multi-stop route for 50 deliveries")

We are also proud of building a robust, enterprise-ready, and stateless architecture. This isn't just a prototype; it's a scalable platform that can serve a global user base securely and efficiently. Finally, we're proud that we've built a future-proof foundation that directly addresses the growing need to connect AI and LLMs to the real world.

What we learned

Throughout this process, we learned that the biggest value we can provide to the developer community isn't just code, but clarity and abstraction. The real "product" is the reduction of cognitive load for developers.

We learned that a well-designed API orchestration layer is exponentially more valuable than a simple API wrapper. By intelligently managing which underlying services are called, we can deliver significant cost savings and performance improvements to the end-user.

Most importantly, we learned that the future of location intelligence lies in its seamless integration with AI. By building GeoServi Core as a "tooling protocol," we've positioned ourselves and our clients at the forefront of this new paradigm.

What's next for GeoServi Core

This is just the beginning. Our vision for GeoServi Core is to become the definitive platform for building location-aware applications. Our roadmap includes:

  • Expanding the Toolset: Integrating other powerful Google APIs, such as the Air Quality API and Solar API, into our set of standardized tools.
  • Developing Industry-Specific Agents: Creating specialized "Client MCPs" pre-trained for the unique needs of industries like Logistics (First-Mile Optimization), Finance (Geo-Risk Analysis), and Retail (Proximity Marketing).
  • Launching as a full-fledged SaaS Product: Making GeoServi Core publicly available to empower the global developer community, from individual hobbyists to large enterprises, to build the next generation of geospatial applications.

Built With

Share this project:

Updates