Role: Python Engineer – Agentic AI & MCP Orchestration
Location: Jersey City, NJ/ Pennington, NJ
FTE only
Job Description
Must Have Technical/Functional Skills
Technical Skills
• Strong hands-on experience in Python for building production-grade applications.
• Experience developing agentic AI applications (multi-agent workflows, tool-using agents, autonomous task execution).
• Expertise with AI orchestration frameworks (agents, tools, planners, workflow controllers).
• Practical experience setting up and configuring Model Context Protocol (MCP) servers.
• Ability to implement and integrate MCP Clients with external systems and AI models.
• Proficiency working with LLMs, prompt engineering patterns, and structured output handling.
• Experience with API integration, event-driven interactions, and tool/skill registration for agents.
• Strong understanding of asynchronous Python (asyncio, concurrency patterns).
• Familiarity with vector stores, embeddings, RAG pipelines, and memory architectures (nice to have).
• Experience working with CI/CD, Git, testing frameworks (pytest), and secure coding practices.
Functional Skills
• Ability to translate business problems into agent-driven automation workflows.
• Strong debugging and troubleshooting of distributed agent behavior and orchestration flows.
• Familiarity with governance, model safety constraints, and responsible AI usage patterns.
• Strong documentation habits for API schemas, MCP interface definitions, and agent lifecycle behavior.
• Effective communication with architects, product teams, and business stakeholders.
• Comfort working in Agile environments with rapid experimentation and iteration.
Roles Responsibilities
Design and develop Python-based agentic applications capable of orchestrating autonomous and semi-autonomous workflows.
• Build, configure, and maintain MCP servers to expose tools, data sources, or domain capabilities to LLM agents.
• Implement and configure MCP clients to interact with multiple MCP tools, AI models, and external systems.
• Develop orchestration logic to coordinate multi-agent behaviors, tool execution, validations, and decision routing.
• Integrate LLMs with internal systems using structured prompts, tool definitions, and safe execution patterns.
• Optimize agent workflows for reliability, performance, security, and cost.
• Create reusable frameworks for agent tools, context handling, memory, and reasoning cycles.
• Collaborate with architecture, platform, and product teams to align on engineering best practices.
• Implement observability: logging, tracing, and monitoring of agent reasoning steps and tool calls.
• Document MCP schemas, agent behaviors, tool interfaces, and operability guidelines.