Software Developer -MCP and Agentic Tooling
Location: Remote
Required Clearance: Public Trust
ManpowerGroup Public Sector (MGPS) provides Skilled Experts and Leading-Edge Solutions to Support Our Nation’s Most Critical Intelligence, Defense and Civilian Missions.
U.S. Citizenship Mandatory: Due to our US federal government contract, candidates for this position are required to be a US Citizen and will be subject to a background investigation.
Job Description: We're looking for a software developer to design and build tools and integrations using the Model Context Protocol (MCP), enabling AI agents to securely and reliably interact with internal systems, data sources, and workflows. This role sits at the intersection of traditional software engineering and emerging agentic AI infrastructure — you'll build the "hands" that let LLM-based agents take real action and access real data. We recognize this is a new and specialized space. We're prioritizing strong software engineering fundamentals and a demonstrated ability to learn protocol-level and AI-specific concepts quickly over exact prior MCP experience.
Job Responsibilities:
- Design, build, and maintain MCP servers that expose internal tools, APIs, and data sources to AI agents.
- Write clear, well-scoped tool and resource definitions (including schemas and descriptions) that LLMs can reliably interpret and use.
- Integrate MCP tools with internal data systems (databases, data warehouses, APIs), ensuring correct, secure, and performant data access.
- Debug and test agentic tool-calling behavior — not just "does the code run" but "does the model use this tool correctly.”
- Use AI coding tools (e.g., Claude Code) as part of your day-to-day development workflow — for scaffolding, debugging, code review, and accelerating delivery — while maintaining ownership of code quality and architectural decisions.
- Collaborate with stakeholders across engineering, data, and business teams to identify workflows worth exposing to agents.
- Document tools and integrations for both human developers and downstream AI consumers.
- Stay current on MCP, agent frameworks, and the broader LLM tooling ecosystem as it evolves.
Required Skills:
- Phython – Advanced
- Database Technologies – Intermediate
- Anthropic Claude Code – Intermediate
- Github – Intermediate
Education Level: Bachelor’s
MGPS specializes in solving our clients’ diverse linguistic, technical and data management challenges created by the dramatic rise in the volume, variety and complexity of open source data. We harness machine learning, artificial intelligence and the latest data acquisition and mining tools, paired with expert human skills to provide customized solutions for every customer.