Graduation Project: Agentic Edge AI

Context-aware LLM decoding for safe and secure Agentic Edge AI

Are you inspired by machine learning, data, AI, currently pursuing your studies, and ready for a new challenge by Context-aware LLM decoding for safe and secure Agentic Edge AI?

Do you want to contribute to “real world” applications and gain experience developing innovative ideas?

Then you will want to be a part of the rising Artificial Intelligence Competence Center at NXP, a leading semiconductor company.

Responsibilities:

At the AI Competence Center, we are looking for a student passionate about developing context-aware decoding strategies for large language models (LLMs) that enable safe and secure autonomous behavior in edge AI agents.

Leverage your creativity for implementing new, fun, and innovating ways of LLM decoding and agentic reasoning in one of our lighthouse projects.

During this internship, our team members will support you to:

• Explore and implement decoding strategies that incorporate contextual signals (e.g., sensor data, user intent, environmental constraints).
• Evaluate safety and security metrics and guardrails for LLM-based agents operating on edge devices.
• Prototype and benchmark agentic behaviors using NXP’s edge AI platforms.

In the project, you will learn to push your software development and machine learning skills further.

You will invent, implement, and integrate a context-aware decoding module for LLMs that enhances agentic decision-making at the edge.

There is no need to start from scratch; instead, you will bring our existing edge AI infrastructure further with your own ideas.

After your internship, you and the team will be proud of having developed a robust and secure LLM-based agentic framework for edge deployment.

Preferred Skills:

• Strong foundation in machine learning and natural language processing
• Experience with LLMs and decoding techniques (e.g., beam search, sampling, constrained decoding)
• Familiarity with edge computing platforms and resource-aware AI
• Interest in safety, security, and trustworthiness in AI systems
• Programming skills in Python and/or C++
• Ability to work independently and collaboratively in a research-driven environment

Your Profile:

• Currently enrolled in a Master’s program in Computer Science, AI, Data Science, or related field
• Passionate about cutting-edge AI research and real-world applications
• Curious, proactive, and eager to learn
• Fluent in English (spoken and written)