About the role
from listingAt American Express, we empower future technologists and researchers to learn, experiment, and make an impact from day one. As a Master’s-level AI Engineer Intern focused on AI Security Research in Enterprise Technology Services, you’ll join a 16-week Summer Internship Program and contribute to hands-on projects at the intersection of artificial intelligence, cybersecurity, and software engineering.
You’ll apply graduate-level coursework or research experience to investigate emerging technologies, build working prototypes and proof-of-concepts, evaluate technical approaches, and help translate research ideas into measurable enterprise outcomes.
In this role, you may explore AI security, agentic AI, application security, generative AI and advanced AI models for cybersecurity, security automation, cloud security, secure software development, and other emerging technologies. You’ll work with AI engineers, cybersecurity engineers, researchers, architects, product partners, and business stakeholders to turn research questions into code, experiments, evaluation results, and reusable capabilities while applying responsible AI, security, privacy, and governance practices.
Enterprise Technology Services teams build and operate technology that helps American Express deliver trusted, secure, and customer-first products and services. The AI Security Research team explores emerging AI and cybersecurity capabilities, develops prototypes and experimental environments, and evaluates how new technologies may be applied responsibly and securely in an enterprise setting.
As a Master’s-level AI Engineer Intern, you’ll contribute with guidance from experienced engineers and researchers while taking ownership of defined research and engineering workstreams. You’ll gain practical experience framing technical questions, designing experiments, building and testing solutions, analyzing results, and communicating findings to technical and non-technical partners. You can expect to:
Research emerging artificial intelligence and cybersecurity technologies, threats, techniques, and security capabilities; synthesize technical literature and translate findings into testable research questions.
Design and build prototypes, proof-of-concepts, security test harnesses, and experimental environments using Python and other modern programming languages.
Develop and test AI and agentic AI solutions, including LLM integrations, AI agents, tool calling, orchestration frameworks, retrieval patterns, and automated workflows.
Experiment with AI models for cybersecurity and software engineering use cases such as security analysis, threat detection, vulnerability research, application security testing, and automation.
Integrate APIs, AI models, security tools, cloud services, containers, data sources, and development environments to support research, prototyping, and technical evaluation.
Design test cases, benchmarks, and controlled experiments to evaluate solution effectiveness, reliability, security, limitations, and potential enterprise applications.
Conduct hands-on evaluations of open-source technologies, products, platforms, and emerging technical capabilities, validating technical claims through reproducible testing and analysis.
Analyze experimental results using qualitative and quantitative evidence; document findings, assumptions, risks, technical limitations, and recommendations while following responsible AI, security, privacy, governance, and compliance expectations.
Prepare technical documentation, research summaries, demonstrations, and presentations; share findings with AI, cybersecurity, engineering, architecture, product, and business stakeholders; and participate in Agile team practices and code reviews.
Minimum Qualifications
Currently enrolled in a full-time Master’s degree program in Computer Science, Artificial Intelligence, Machine Learning, Cybersecurity, Computer Engineering, Software Engineering, Data Science, Information Systems, or another related technical field.
Students must have a graduation date aligned to the 2027 Graduate Masters Summer Internship eligibility window.
Strong hands-on programming experience, preferably with Python, and exposure to one or more languages such as Java, JavaScript/Node.js, Go, C/C++, Bash, PowerShell, or similar technologies.
Solid understanding of computer science and software engineering concepts, including data structures, algorithms, APIs, debugging, testing, version control, software design, and problem solving.
Graduate-level understanding of artificial intelligence, machine learning, generative AI, LLMs, AI agents, model evaluation, or related concepts through coursework, thesis work, research, projects, or labs.
Understanding of cybersecurity principles, threats, vulnerabilities, security controls, application security, cloud security, threat modeling, or secure software development practices.
Ability to independently investigate a bounded technical problem, design an experimental approach, build or test working solutions, analyze evidence, and document reproducible findings with appropriate guidance.
Strong analytical, problem-solving, written and verbal communication, collaboration, curiosity, and learning agility with the ability to work effectively in a research and engineering team environment.
Preferred Qualifications
Demonstrated hands-on experience through graduate coursework, thesis or academic research, personal projects, open-source contributions, security labs, hackathons, Capture-the-Flag competitions, internships, or other technical work.
Interest in the intersection of artificial intelligence, cybersecurity, software engineering, and emerging technologies, including AI security, agent security, application security, cloud security, security automation, or secure AI-enabled software development.
Experience building software projects, research prototypes, proof-of-concepts, experiments, test harnesses, intelligent assistants, AI agents, agentic workflows, or applied research demonstrations.
Experience with LLMs, generative AI, prompt engineering, embeddings, retrieval-augmented generation, tool or function calling, orchestration frameworks, model evaluation, or human-in-the-loop review.
Exposure to security research or testing techniques such as threat modeling, vulnerability research, AI red teaming, automated security testing, agent security testing, adversarial evaluation, or evaluation of AI-native cybersecurity tools.
Familiarity with APIs, cloud platforms, containers, data pipelines, development environments, security tools, or software integration patterns.
Experience designing benchmarks, experiments, or technical evaluations and using evidence to assess effectiveness, limitations, risks, and recommendations.
Awareness of secure software development, responsible AI, data protection, privacy, governance, compliance, research ethics, and responsible handling of sensitive information.
Strong initiative, responsible judgment, attention to detail, presentation skills, and comfort working collaboratively in an experimental, fast-learning environment.
Employment eligibility to work with American Express in Canada is required as the company will not pursue visa sponsorship for these positions.