AI Research Engineer - NLP

We have an exciting opportunity for an AI Research Engineer at UL Research Institutes, within our Digital Safety Research Institute (DSRI). This is a REMOTE opportunity

The AI Research Engineer advances Digital Safety mission by designing, developing, and evaluating advanced artificial intelligence (AI) safety assessment systems, with a focus on large language models (LLMs), natural language processing (NLP) safety. This role contributes to the development of rigorous evaluation methodologies, scalable assessment capabilities, and research-driven insights that improve the safety assessments of frontier AI systems.

UL Research Institutes:

At UL Research Institutes (ULRI), we expand the boundaries of safety science to create a more secure and sustainable world. For more than a century, we have studied the unintended consequences of innovation, designed solutions to mitigate risk and shared our findings with academia, scientists, manufacturers, and policymakers across industries. We identify critical safety and sustainability issues, asking tough questions because we believe a safer world begins with knowledge.

Build a safer, more secure, and sustainable future with us. Join us and work with our Fire Safety teams who conduct the research required to produce that knowledge and put it into practice.

Digital Safety Research Institute:

The Digital Safety Research Institute (DSRI) addresses unresolved safety risks and emerging dangers in the world’s digital ecosystem. We aim to help protect people from unwanted access to personal devices and privacy, algorithms, and disinformation threats by working to create a new digital safety ecosystem.

DSRI is committed to sharing digital safety insights, collaborating with key stakeholders, and offering extensive outreach and education to help support safer digital environments for people everywhere.

What you’ll learn and achieve:

As the AI Research Engineer, you will play a key role in the rapid growth of UL as you:

• Studies incident reports implicating LLMs or NLP algorithms and hypothesize root causes.

• Conducts literature and open-source reviews in NLP or LLM Designs, plans, and creates NLP or LLMs assessments - expanding digital safety best practices. Applies technical, diagnostic, and troubleshooting skills.

• Codes or trains text analytics needed to run NLP or LLM assessments. Prepares written component documentation for technical audiences.

• Publishes assessment research results, on-line, using Digital Safety lab tools, and in peer-reviewed conferences and journals.

• Works closely with team members to ensure that all processes are accurate and technically sound.

What makes you a great fit:

While no one candidate will embody every quality, the successful candidate will bring many of the following professional competencies and personal attributes:

• Demonstrates mastery level knowledge of NLP and LLM safety and how safety assessments can protect the public from LLM harms.

• Solves AI assessment research challenges independently and efficiently.

• Engages software engineering teams who build tools to help DSRI researchers scale their assessments to perform rapidly at scale

• Looks ahead to needed research and development next steps, for better analytics to assess LLM or NLP outputs.

• Begins developing relationships within the broader safety community

Professional education and experience requirements for the role include:

• Ph.D. in computer science or related field, focused on NLP or LLMs.

• 3 years of relevant experience, post Ph.D.

About UL Research Institutes and UL Standards & Engagement