Postdoctoral Researcher: AI Safety & Model Editing

University of Virginia

Charlottesville (VA)

On-site

USD 60,000 - 75,000

Full time

3 days ago
Be an early applicant

Get a reply from this employer

Your resume and cover letter, tailored to exactly what they’re hiring for.

Get past ATS filters

Benefits offered by this job

Health insurance
Vision Coverage
Dental Plan
Benefit Savings Plans
Life Insurance
Disability Benefits
Paid Time Off: 22 days + holidays + 8w

Job summary

The University of Virginia is seeking a research professional to develop and evaluate machine unlearning and model editing methods aimed at reducing hazardous biological capabilities in AI systems while preserving beneficial scientific functions.

The role reports to Assistant Professor Tom Hartvigsen and collaborates with SDS faculty and external partners to build an open evaluation suite for AI biosecurity.

Qualifications

  • PhD or equivalent in data science, CS, ML, or related field.
  • Proficient Python; hands-on with PyTorch and Hugging Face Transformers.
  • Strong publication record in ML, NLP, and biosecurity.
  • Experience training/finetuning/post-training for large language models.
  • Software engineering practices to support reproducible research.

Responsibilities

  • Implement and compare machine unlearning and model editing methods, including gradient-based fine-tuning and inference-time steering.
  • Design and run experiments to measure safety-utility Tradeoffs and benchmark performance.
  • Develop adversarial testing protocols with partners, including jailbreaks and ensemble attacks.
  • Lead development of the open-source UBS-Bio evaluation suite with baselines and docs.
  • Support interpretability analyses to identify hazardous vs beneficial representations.
  • Prepare manuscripts and publish and maintain reproducible code releases.

Skills

Python
PyTorch
Hugging Face Transformers
Publications
LLM fine-tuning

Education

PhD in data science, computer science, ML, or related field

Job description

The University of Virginia is seeking a research professional to develop and evaluate machine unlearning and model editing methods aimed at reducing hazardous biological capabilities in AI systems while preserving beneficial scientific functions.

The role reports to Assistant Professor Tom Hartvigsen and collaborates with SDS faculty and external partners to build an open evaluation suite for AI biosecurity.

Get your free, confidential resume review.

or drag and drop your file here.

Similar jobs

Similar jobs worth comparing

Postdoctoral Research Associate, Machine Unlearning and Model Editing for AI Biosecurity
Postdoctoral Research Associate, Machine Unlearning and Model Editing for AI Biosecurity

University of Virginia • Charlottesville (VA)

On-site
USD 60,000 - 75,000
Health insurance
Vision Coverage
Dental Plan
+4
Senior AI Safety Scientist — Biology Risk (Remote US)
Senior AI Safety Scientist — Biology Risk (Remote US)

SecureBio, LLC • United States

Remote
USD 235,000 - 300,000
Healthcare coverage
Flexible work hours
Conference sponsorships
+1
BioSafety AI Research Scientist — Safety & Evaluation
BioSafety AI Research Scientist — Safety & Evaluation

Anthropic Limited • San Francisco (CA)

On-site
USD 300,000 - 405,000
AI Biology Safety Scientist & Model-Risk SME
AI Biology Safety Scientist & Model-Risk SME

Google LLC • Mountain View (CA)

On-site
USD 174,000 - 252,000
Research Scientist: Harmful Manipulation & AI Safety
Research Scientist: Harmful Manipulation & AI Safety

Google LLC • New York (NY)

On-site
USD 174,000 - 252,000
AI Safety & Privacy Research Scientist
AI Safety & Privacy Research Scientist

Meta • San Francisco (CA)

On-site
USD 122,000 - 181,000
Bonus
Equity
Benefits
Postdoctoral Researcher: AI-Driven Cybersecurity
Postdoctoral Researcher: AI-Driven Cybersecurity

Virginia Commonwealth University - College of Engineering • Richmond (VA)

On-site
USD 50,000 - 60,000
AI Safety Researcher: Red-Teaming, Safe RLHF, Unlimited PTO
AI Safety Researcher: Red-Teaming, Safe RLHF, Unlimited PTO

Bison • San Francisco (CA)

On-site
USD 350,000 - 475,000
Health benefits
Dental benefits
Vision benefits
+3
AI Safety Researcher - Inside Risk & Model Evaluation
AI Safety Researcher - Inside Risk & Model Evaluation

TypeSafe AI • San Francisco (CA)

On-site
USD 150,000 - 210,000
Senior AI Biosecurity Scientist — Remote, Flexible PTO
Senior AI Biosecurity Scientist — Remote, Flexible PTO

SecureBio, LLC • Cambridge (MA), Berkeley (CA)

Hybrid
USD 235,000 - 300,000
Flexible work hours
Unlimited paid time off (PTO)
Parental leave (12 weeks)