Agentic AI MCP Integration Specialist

Specializing in extending agents into enterprise systems through MCP (Model Context Protocol). Build MCP servers, integrate SaaS/SOR APIs, and harden the tool layer that agents depend on.

• 5+ years of experience in software engineering, API integration, data engineering, or enterprise application integration.
• Hands‑on experience with Generative AI / LLM-based applications.
• Experience building APIs, connectors, microservices, or enterprise integration layers.
• Exposure to Agentic AI, tool‑calling, workflow automation, or AI orchestration frameworks.
• Bachelor's or Master's degree in Computer Science, Information Technology, Engineering, Data Science, or related field.

• Design, build, and maintain MCP servers that expose enterprise tools, data, and APIs to AI agents.
• Develop reusable MCP connectors/adapters for internal and external systems.
• Implement MCP-based access to:

• Databases and data platforms
• REST / GraphQL APIs
• Document repositories
• Workflow systems
• Enterprise applications

• Ensure MCP servers are modular, reusable, secure, and observable.
• Integrate MCP‑enabled tools with LLM applications, copilots, and autonomous agents.
• Implement tool invocation patterns for single‑agent and multi‑agent workflows.
• Ensure AI agents can Client, invoke, and consume tools safely through MCP.
• Build integrations between Agentic AI platforms and enterprise systems including:

• Service Now, Jira, Confluence
• SharePoint, Teams, Outlook
• Databricks, Snowflake, Synapse
• SQL / No
  
  SQL databases
• CRM, ERP, and insurance platforms
• Internal APIs and microservices

• Work with application, data, cloud, and cybersecurity teams to onboard enterprise tools into MCP‑based agent ecosystems.

Design and implement MCP servers exposing enterprise tools (Rally, Service Now, Confluence, SharePoint, Git Hub Jenkins, Snowflake, etc.).

Define tool schemas, resource URIs, and prompt templates following the MCP spec; ensure deterministic, idempotent, well‑documented tools.

Implement authentication patterns (OAuth2, SSO, AWS IAM, service principals) and per‑tenant authorization for MCP tools.

Build connectors and adapters that translate enterprise API quirks (pagination, rate limits, eventual consistency) into clean agent‑facing primitives.

Instrument MCP servers with logging, tracing, metrics, and audit trails for compliance.

Partner with the Architect on the MCP tool catalog, naming conventions, and versioning strategy.

Triage and resolve issues where agents misuse tools — improve schemas, descriptions, and error responses.

Deep understanding of the MCP specification (resources, tools, prompts, transports — /HTTP/SSE).

Strong API integration background: REST, webhooks, OAuth2/OIDC.

Python and/or Type Script proficiency; familiarity with MCP SDKs in either ecosystem.

Enterprise integration patterns: retries, circuit breakers, backoff, dead‑letter queues.

Security: secrets vaulting (AWS Secrets Manager / Hashi Corp Vault), least‑privilege scoping.