To apply, in addition to your application on the careers.umich.edu website, please email the following to drmalik@umich.edu with the subject line: RA -Neurosymbolic Web Filtering (SMILES Lab)

• Resume or CV
• Links to relevant code repositories, projects, or previous research 

The Secure Modeling and Intelligent Learning Systems (SMILES) Lab at the University of Michigan-Flint is hiring a part-time research assistant for a temporary, hourly-paid position lasting up to three months. The project focuses on developing neurosymbolic multimodal web filtering systems, combining neural models with symbolic reasoning to detect, analyze, and filter content across text, images, video, and audio streams. 

This role is open to both UM-Flint students and external candidates who are senior undergraduate, graduate, or postdoc students, including those interested in gaining experience in AI and cybersecurity.

• Design and implement machine learning techniques for web content classification and filtering, including handling encrypted DNS traffic (e.g., DoH, DoT).
• Build and refine knowledge graphs from multimodal data using large language models (LLMs) to enable structured reasoning and semantic understanding.
• Develop and integrate retrieval-augmented generation (RAG) pipelines to enhance contextual comprehension across modalities.
• Curate, preprocess, and manage large-scale multimodal datasets for training, fine-tuning, and evaluation.
• Participate in weekly lab meetings and contribute to collaborative publications and research outputs in the field of multimodal AI.

• Must be a senior undergraduate, graduate, or postdoc student
• Extensive programming experience in Python, with proficiency in Torch, Scapy, C programming, and Linux command-line tools.
• In-depth knowledge of machine-learning, data science, networking, web applications, or operating systems 
• Technical writing and communication
• Self-motivated, detail-oriented, and capable of working both independently and within a team.

• Strong passion for programming 

Hybrid
The work requirements allow both onsite and offsite work and an employee has an expected recurring onsite presence. On occasion, the employee may be required and must be available to work onsite more frequently if necessitated by unit leadership or their designee and/or the job requirements.

Positions that are eligible for hybrid or mobile/remote work mode are at the discretion of the hiring department. Work agreements are reviewed annually at a minimum and are subject to change at any time, and for any reason, throughout the course of employment. Learn more about the work modes.

Job openings are posted for a minimum of three calendar days.  The review and selection process may begin as early as the fourth day after posting. This opening may be removed from posting boards and filled anytime after the minimum posting period has ended.

The University of Michigan is an equal employment opportunity employer.