Noblis is looking to solve real-world Federal missions and contribute to making AI systems safer, more transparent, and better understood through advanced research on LLMs
Requirements
- Familiarity with modern machine learning frameworks such as PyTorch, TensorFlow, JAX, or similar tools
- Comfortable with modern software development practices including version control systems, automated testing, and debugging
- Hands-on experience working with large neural models
- Prior experience in a research lab or academic setting, particularly in machine learning, NLP, or AI
- Prior publications at top conferences (NeurIPS, ICML, ICLR, AAAI, CVPR, ICRA, etc.)
Responsibilities
- Conduct research on advanced LLM architectures, interpretability, alignment, and safety
- Design and execute experiments exploring behavior of LLMs at scale, e.g. representation analysis, feature attribution, circuit-level understanding, or automated interpretability techniques
- Collaborate with experienced researchers to develop innovative approaches that bridge theoretical research and practical AI challenges
- Communicate findings through presentations and written reports, contributing to broader organizational research objectives
Other
- US Citizenship
- Currently pursuing a bachelor's degree or master's degree in computer science or related disciplines with a 3.3 GPA with a graduation date in 2026 or 2027
- Ability to work up to 20 hours a week during the school year and transition to 10 week in person experience in the summer
- Motivation to advance research with real-world impact to deepen understanding and improve transparency of LLMs
- Proven ability to independently pursue research, proactively contribute innovative ideas, and thrive within collaborative teams
