Agentic AI Engineering Intern

Leidos is a technology leader serving government and commercial customers with innovative digital solutions. They are seeking an Agentic AI Engineering Intern to gain hands-on experience in developing generative AI solutions that enhance decision intelligence across mission-critical applications.

Responsibilities:

• Assist in the development and testing of agentic AI systems, including Multi-Agent and Agent-to-Agent (A2A) workflows, leveraging common industry standards such as the Model Context Protocol (MCP) to create interoperable AI agents
• Support the implementation of MCP Tools and Resources that enable Large Language Models (LLMs) to interact with internal systems and APIs in a secure, standardized manner
• Collaborate with engineers and data scientists to contribute to the architecture of a centralized "AI Gateway" that provides a unified, platform-independent interface for leveraging various LLMs
• Help implement observability pipelines to track trace-level data, monitor model latency, and support the optimization of Generative AI systems in production
• Work closely with senior team members to translate strategic designs into functional, production-ready solution components
• Participate in the implementation of AI guardrails to filter inputs and outputs, supporting data security, integrity, and the prevention of adversarial attacks such as prompt injection
• Assist in the design and implementation of Retrieval-Augmented Generation (RAG) pipelines to enhance LLM accuracy and grounding with enterprise data sources
• Learn and apply engineering best practices including version control (Git), automated testing, and CI/CD processes for AI systems
• Stay current with emerging trends in agentic AI, operational AI, and MLOps, and contribute ideas to continuously evolve the team's capabilities

Requirements:

• Currently pursuing a Bachelor's degree (rising junior or senior preferred) in Computer Science, Artificial Intelligence/Machine Learning, Engineering, or a closely related quantitative field
• US citizenship required
• Proficiency in Python and familiarity with at least one major ML library or framework (e.g., TensorFlow, PyTorch, Scikit-learn, or Hugging Face Transformers)
• Basic understanding of the machine learning lifecycle, including data preparation, model training, evaluation, and deployment concepts
• Demonstrated interest in agentic AI patterns, multi-agent systems, and/or LLM-based workflows (e.g., through coursework, personal projects, or research)
• Foundational understanding of cybersecurity principles as they relate to AI systems
• Familiarity with version control systems (e.g., Git/GitHub)
• Strong analytical, problem-solving, and communication skills
• Ability to work collaboratively in a team-oriented environment
• Experience with Retrieval-Augmented Generation (RAG) architectures or vector database technologies (e.g., Pinecone, Weaviate, ChromaDB, FAISS)
• Exposure to MLOps platforms or workflows (e.g., MLflow, Kubeflow, or cloud-native ML services)
• Familiarity with containerization technologies (e.g., Docker) and basic orchestration concepts (e.g., Kubernetes)
• Hands-on experience with a major cloud platform (AWS, Azure, or GCP)
• Experience with or coursework in prompt engineering, LLM fine-tuning, or agent framework development (e.g., LangChain, LangGraph, CrewAI, AutoGen)
• Familiarity with the Model Context Protocol (MCP) or similar standards for LLM tool integration
• Knowledge of AI ethics, responsible AI practices, or federal AI compliance standards (e.g., NIST AI RMF)
• Knowledge of AI security frameworks such as MITRE ATLAS
• Contributions to open-source AI/ML projects or a public portfolio of AI-related work (e.g., GitHub, Kaggle)
• Prior internship, co-op, or research experience in a national security, defense, or intelligence community environment