Experience:

• 3-5 years
• 2–4 years in software or AI/ML engineering
• At least 1 year on agentic or LLM-powered systems

Responsibilities:

• Develop and maintain agentic pipelines including tool-use chains, ReAct loops, and stateful multi-step reasoning workflows
• Build and extend Agent Harness components including tool schemas, agent personas, execution environments, and guardrail configurations
• Implement and maintain MCP server and client integrations for structured tool invocation
• Develop Agent-to-Agent communication flows including task delegation, shared context management, and messaging protocols
• Assist in fine-tuning language models using domain-specific datasets
• Support training pipeline setup and evaluation benchmarking
• Integrate Claude SDK and other LLM provider APIs with prompt engineering and response handling
• Build user-facing interfaces with live streaming outputs and task dashboards using React and HTMX
• Participate in architecture reviews, testing, and documentation of agent interaction maps and system behaviors

Additional responsibilities:

• Assist company from time to time

Nice to have:

• Basic TypeScript knowledge

More skills:

agentic pipelines, Model Context Protocol (MCP), Agent-to-Agent (A2A) communication, language model fine-tuning, Qdrant, Pinecone, Chroma, Python, TypeScript

Other:

• EC-Council is the world's largest cyber security technical certification body operating in 145 countries
• Equal opportunity workplace and affirmative action employer
• Committed to providing reasonable accommodations to individuals with disabilities
• Candidates must have legal authorization to work in the country of the position
• Work submitted during recruitment becomes exclusive property of EC-Council
• Privacy policy governs data collection and handling during recruitment