Hero Image

AnitaB.org Talent Network

Connecting women in tech with the best professional opportunities!

POST-DOCTORAL FELLOWSHIP IN AGENTIC RED TEAMING OF AI MODELS

Carnegie Mellon University

Carnegie Mellon University

Software Engineering, Data Science
Pittsburgh, PA, USA
Posted on Mar 10, 2026

POST-DOCTORAL FELLOWSHIP IN AGENTIC RED TEAMING OF AI MODELS

4.2 de 5 estrellas
Pittsburgh, PA

Descripción completa del empleo

Carnegie Mellon University: Heinz College

Location

Pittsburgh, PA

Open Date

Mar 03, 2026

Description

The Heinz College of Information Systems and Public Policy at Carnegie Mellon University invites applications for a Postdoctoral Fellowship focused on the red teaming of AI systems. This position is part of the AI Measurement Science and Engineering Center (AIMSEC), a university-wide initiative dedicated to advancing rigorous methods for evaluating, stress-testing, and securing AI systems.

We seek a scholar interested in developing next-generation methods for agentic red teaming — including reinforcement learning–based adversarial policy generation, automated vulnerability discovery, and modeling of human red teaming behaviors in networked and cyber-physical systems. The fellow will work with an interdisciplinary group of faculty spanning computer science, public policy, systems engineering, and behavioral science.

Areas of interest and emphasis include:

  • Reinforcement learning for automated adversarial testing
  • Multi-agent and adversarial learning frameworks
  • Red teaming of foundation models and networked AI systems
  • Human-in-the-loop adversarial modeling
The fellow will have the opportunity to define independent research directions, lead publications, and collaborate across the CMU ecosystem.
The Heinz College is a multidisciplinary academic unit focused on problem-driven research addressing societal challenges. AIMSEC brings together faculty across CMU to advance the science of AI evaluation and robustness. The postdoctoral fellow will engage with researchers across engineering, computer science, and policy.

Qualifications

Applicants should have received a Ph.D. within the past year (or expect to receive one by September 2026) in Computer Science, Machine Learning, Electrical Engineering, Operations Research, or a related field. Successful candidates will demonstrate:

  • Strong technical background in reinforcement learning
  • Evidence of research in adversarial ML, multi-agent systems, cyber-physical systems, or related domains
  • A publication record in leading venues
  • Interest in real-world AI safety and security challenges
  • Value working in an inter-disciplinary environment

Application Instructions

Applications should include a cover letter, a C.V., up to two research papers, and contact information for three professional references. In their cover letter, applicants are invited to describe how they would be a particularly good fit for Heinz College and this opportunity. Applications will be reviewed beginning April 1, 2026. Applications will be accepted until the position is filled.

Application Process

This institution is using Interfolio's Faculty Search to conduct this search. Applicants to this position receive a free Dossier account and can send all application materials, including confidential letters of recommendation, free of charge.
crea una cuenta de Indeed antes de continuar a la página web de la empresa.