We are working with a US-based client in the veterinary technology and pet healthcare industry. The client provides digital solutions that connect veterinary practices, pet owners, pharmacy services, and business partners.

We are helping the client progress through an AI transformation, including establishing AI foundations, governance practices, agentic workflows, and scalable approaches to AI product development.

As part of this transformation, the client has been evaluating a third-party AI/MCP gateway and would like to determine whether a comparable solution can be built using Google Cloud technologies. The project will involve reviewing an existing proof of concept, designing and building a Google-based alternative, and comparing the two approaches across functionality, architecture, security, observability, scalability, operational complexity, and cost.

This is a technically focused proof-of-concept engagement that requires limited veterinary-domain knowledge. The outcome will help the client select the appropriate platform and architecture for future AI products.

The ideal candidate combines strong hands-on AI engineering expertise with solution architecture skills. They should be able to work independently, turn a high-level concept into a functioning prototype, identify architectural gaps, communicate trade-offs, and provide an objective, evidence-based platform recommendation.

• Review the existing AI/MCP solution and understand its primary capabilities.

• Gather and document technical and non-functional requirements.

• Design a comparable architecture using Vertex AI and other relevant Google Cloud services.

• Build a working proof of concept for selected agentic workflows.

• Develop AI agents that connect to internal data, APIs, and business tools.

• Implement or integrate MCP servers and tool-calling workflows.

• Configure authentication, authorization, secrets management, and data-protection controls.

• Add logging, tracing, monitoring, model evaluation, and cost tracking.

• Identify capabilities that Vertex AI cannot provide natively and recommend supporting services.

• Compare the existing and Google-based solutions based on:

  • Functionality and limitations.

  • Implementation and integration effort.

  • Security and governance.

  • Reliability and scalability.

  • Observability and maintainability.

  • Infrastructure and model costs.

• Document the architecture, technical findings, risks, and recommendations.

• Present the proof of concept and support its handover to the client’s engineering team.

• Strong practical experience with Google Cloud Platform.

• Hands-on experience with Vertex AI and Gemini models.

• Experience building AI agents, tool-based workflows, or multi-step AI applications.

• Knowledge of Vertex AI Agent Builder or Google’s enterprise agent tooling.

• Experience integrating LLM applications with APIs, databases, and enterprise systems.

• Understanding of MCP or similar protocols for connecting AI agents to tools and data.

• Strong Python or TypeScript skills.

• Knowledge of prompt design, structured outputs, retrieval-augmented generation, and model evaluation.

• Experience implementing AI security controls, including IAM, secrets management, PII protection, and secure data handling.

• Familiarity with cloud observability, logging, tracing, and cost monitoring.

• Ability to evaluate competing technologies and communicate technical trade-offs clearly.

• Strong written and spoken English for direct collaboration with client stakeholders.