Internship / Master Thesis: Securing the Future of Agentic AI

ELCA; Schweiz; Hybrid; CHF 17.000 - 23.000; Vollzeit

Benefits dieser Stelle

2 days per week from home

Monthly After-Works

Zusammenfassung

ELCA in Switzerland offers a 6-month internship or master thesis starting February 2027, focusing on agentic AI security. You will study secure architectures for autonomous AI agents and contribute to a practice at the intersection of AI innovation and enterprise security.

You will conduct literature reviews, build risk taxonomies, define design principles, implement reference architectures, run defense benchmarks, and deliver a final report and presentation to stakeholders.

Qualifikationen

• Pursuing a master's degree in computer science or related field.
• Strong cybersecurity knowledge and AI system understanding.
• Ability to prototype and automate with code.

Aufgaben

• Literature review and threat landscape mapping.
• Build a taxonomy of agentic AI risks.
• Define security design principles and reference architectures.
• Design and implement secure-by-design agentic architectures.
• Benchmark the effectiveness of defenses against threat scenarios.
• Deliver a final report and internal presentation.

Kenntnisse

Cybersecurity knowledge

AI systems understanding

Threat modelling

Coding prototyping

Ausbildung

Master student in Computer Science or related field

Tools

Python

Jobbeschreibung

Start date: February 2027 · Duration: 6 months
Please include your latest academic transcripts with your application. Applications submitted without transcripts will not be considered.

Capable of autonomously planning, deciding and acting, agentic AI systems are rapidly moving from research labs into production environments. This shift introduces a new class of security challenges: how do you control an AI that controls other systems? How do you audit decisions made autonomously, at scale?

In this 6-month internship or master thesis starting in February 2027, you will conduct applied research on how to build secure agentic AI architectures and help shape the foundation of a practice at the intersection of AI innovation and enterprise security.

Your tasks

• Literature review and threat landscape mapping: survey research on AI agent security, known attack vectors and emerging AI risk frameworks, and identify gaps
• Build a taxonomy of agentic AI risks covering multi-agent architectures, tools, memory mechanisms and autonomous decision chains
• Define security design principles and reference architectures covering sandboxing, access control, auditability and human oversight
• Design and implement blueprints of secure-by-design agentic architectures as reusable reference implementations
• Benchmark the effectiveness of defenses against documented threat scenarios with measurable, reproducible results
• Deliver a final report and internal presentation summarizing findings, methodologies and recommendations

What we offer

• A dynamic, collaborative workplace with a highly motivated, multicultural team working across international sites
• The chance to work on what could be the next generation of architecture
• Good work-life balance (2 days per week from home)
• Monthly After-Works organized at each location

About your profile

• Strong cybersecurity knowledge (authentication, access control and authorization, threat modelling, security by design, etc.)
• General understanding of how AI systems work, ideally with some exposure to large language models and agent-based architectures
• Some proficiency in writing and reading code ( ability to prototype, experiment and automate)