AI Integration Engineer

Spatial Front, Inc. (SFI) is seeking an AI Integration Engineer to support our growing modernization team. SFI was recently awarded the 2025 USA Today National Top Places to Work award and the 2025 Washington Post Top Workplaces. The ideal candidate will perform hands-on development of Model Context Protocol (MCP) servers, tools, connectors, APIs, and supporting integration services that enable artificial intelligence agents to securely interact with enterprise applications, databases, and data sources supporting PeopleSoft HCM and related systems.

This role is focused on building the integration layer between AI agents and enterprise systems. The candidate will design and develop MCP capabilities that expose approved business functions and data to AI applications through secure, controlled, and reusable interfaces. The candidate will work closely with AI developers, PeopleSoft developers, data engineers, DBAs, platform engineers, and cybersecurity personnel to ensure AI agents can access enterprise capabilities without bypassing existing security, authorization, data, or business-process controls. As a valued member of the SFI team, you will help deliver secure and maintainable AI integration capabilities for mission-critical Federal Government systems.

Location

Crystal City, VA - On-Site/Hybrid

Responsibilities

• Design, develop, test, and maintain Model Context Protocol (MCP) servers and supporting integration components using Python.
• Develop reusable MCP tools, resources, prompts, schemas, and interfaces that allow AI agents to interact with approved enterprise applications, databases, APIs, and data sources.
• Build MCP capabilities that securely expose enterprise functions such as search, retrieval, lookup, summarization support, case information, workflow information, knowledge content, and other approved business capabilities.
• Develop Python-based connectors and services that translate MCP tool requests into secure calls to enterprise APIs, databases, web services, and application interfaces.
• Design MCP tools with clearly defined inputs, outputs, validation rules, permissions, error handling, and operational boundaries.
• Implement secure read-only and transactional integration patterns, ensuring AI agents can perform only explicitly authorized actions.
• Develop and maintain REST/SOAP APIs, backend services, adapters, and supporting integration logic where MCP tools require access to existing enterprise systems.
• Integrate MCP services with Oracle databases, enterprise applications, case-management systems, knowledge repositories, and other approved data sources.
• Work with PeopleSoft technical teams to expose approved PeopleSoft HCM and CRM functionality through Integration Broker, Component Interfaces, web services, queries, APIs, database interfaces, or other appropriate integration mechanisms.
• Develop database queries and data-access components used by MCP tools while applying appropriate data filtering, validation, security, and access controls.
• Implement authentication and authorization for MCP servers and downstream enterprise connections using approved service identities, tokens, certificates, credentials, and cloud identity mechanisms.
• Apply least-privilege principles to MCP tools and service accounts so agents receive only the data and capabilities required for a specific use case.
• Implement input validation, output validation, schema enforcement, exception handling, retries, timeouts, logging, and other reliability patterns across MCP and API integrations.
• Prevent inappropriate direct access to enterprise applications or databases by creating controlled integration interfaces and reusable service boundaries.
• Develop mocks, simulated services, test harnesses, and automated integration tests to validate MCP behavior without requiring inappropriate use of sensitive production data.
• Troubleshoot issues across MCP servers, Python code, APIs, authentication, database connections, application interfaces, network connectivity, and downstream services.
• Analyze agent tool calls and integration behavior to identify incorrect tool usage, invalid parameters, excessive permissions, failed transactions, or other integration defects.
• Maintain MCP interface documentation, tool definitions, API specifications, data mappings, authentication requirements, deployment procedures, and operational support documentation.
• Establish reusable MCP development patterns, coding standards, naming conventions, security practices, and integration frameworks for use across multiple AI agents and use cases.
• Participate in code reviews, architecture discussions, backlog refinement, demonstrations, testing, release readiness, and other Agile/SAFe delivery activities.
• Other duties as assigned.