Job Description

Dr. Harry Zhang in the College of Computer & Informatics’ Department of Computer Science seeks a part-time Research Assistant to explore the intersection of Large Language Models (LLMs) and automated planning. The Research Assistant will investigate how LLMs can bridge the gap between natural language requirements and structured formalisms (such as PDDL or temporal logic) to enable autonomous agents to solve complex tasks.

Essential Functions

• Develop and evaluate prompt engineering or fine-tuning strategies to translate ambiguous human instructions into verifiable symbolic representations.
• Conduct benchmarking experiments to assess the accuracy and "hallucination" rates of LLMs when generating domain-specific planning models.
• Collaborate with the engineering team to integrate these formalization pipelines into our existing planning and execution frameworks.

Required Qualifications

• Minimum of a HS Diploma or GED.
• Strong proficiency in Python and experience with LLM APIs (e.g., OpenAI, Anthropic) or open-source frameworks (e.g., Hugging Face, LangChain).
• Solid understanding of symbolic logic or formal methods.

Preferred Qualifications

• Currently pursuing a degree (Master's or PhD preferred) in Computer Science, AI, or a related field.
• Prior experience with Automated Planning (PDDL) or formal verification tools.
• A track record of research contributions or publications in NLP or Robotics.
• Familiarity with neuro-symbolic AI approaches that combine deep learning with classical reasoning.

Location

• Hybrid/Flexible Work Arrangement

Additional Information
This is a Part-Time Temporary position classified as Non-Exempt. The offered pay is based on the successful candidate’s qualifications and experience, department budget, and an internal equity review.

For information about benefits, please review Drexel’s Per Diem/Casual/Temporary Benefit Fact Sheet

Special Instructions to the Applicant
Please make sure you upload your CV/resume and cover letter when submitting your application.

A review of applicants will begin once a suitable candidate pool is identified.

#LI-Remote

#LI-Hybrid