Agentic AI Engineering Intern - Cybersecurity at Qualys · Foster…

Summary: Qualys is a company focused on innovation and teamwork to support exciting missions. They are seeking an Agentic AI Engineering Intern to contribute to building AI-driven cybersecurity products, working closely with the engineering lead to improve the intelligence layer of their AI agent platform. Responsibilities: - Improve and refine tool definitions that AI agents use to interact with security APIs, including descriptions, schemas, and metadata, to ensure consistent and accurate tool selection across the platform - Identify and resolve conflicts or ambiguities across tool definitions that lead to incorrect agent behavior. This is a core part of the agent workflow and will be the primary area of focus - Build evaluation harnesses to measure tool selection accuracy and track improvements over time - Improve the validation layer that determines whether an AI agent’s execution plan is feasible and within its capability boundary. This area works today but needs meaningful improvement - Develop better methods for agents to self-assess plan confidence and determine when to proceed versus when to escalate - Design and implement guardrails that prevent agents from executing outside their defined boundaries - Contribute across the broader agentic AI platform including prompt engineering, search relevance, agent orchestration, and testing - Write tests and build evaluation frameworks for multi-agent workflows - Help improve how the platform matches user intent to agent capabilities using search and retrieval techniques Required Qualifications: - Currently pursuing a BS or MS in Computer Science, AI/ML, Data Science, or a related field - Familiarity with Python - Basic understanding of large language models and generative AI, including how prompting, tool use, and agent loops work - Comfortable navigating and contributing to large codebases - Demonstrates a commitment to continuous learning in this rapidly evolving field Preferred Qualifications: - Exposure to LLM orchestration frameworks or agent-building libraries - Experience with prompt engineering including system prompts, few-shot examples, and chain-of-thought patterns - Familiarity with vector search, embeddings, and retrieval-augmented generation - Understanding of tool-calling protocols and patterns for LLMs - Experience with cloud platforms and containerized development environments - Exposure to web frameworks, backend or frontend Required Skills: Python, Large language models, Generative AI, Prompt engineering, Vector search, Embeddings, Retrieval-augmented generation, Tool-calling protocols for LLMs, Cloud platforms, Containerized development environments, Web frameworks Benefits: Comprehensive and highly competitive benefits package