Postdoctoral Research Scientist

Postdoctoral Research Scientist Columbia University in the City of New York:
The Fu Foundation School of Engineering and Applied Science:
Computer Science

Manhattan, NY

Oct 07, 2026

$75,000-$90,000 annually

The Department of Computer Science at Columbia University is seeking a Postdoctoral Research Scientist to lead research at the intersection of formal methods, computer security, and artificial intelligence. The position focuses on developing formal verification techniques to enforce safety and security properties of autonomous, agentic AI systems and AI swarms.

The successful candidate will work closely with lead faculty to design scalable, agentic verification frameworks capable of leveraging large-scale GPU infrastructure (e.g., Empire AI).

The goal of the project is to automate the specification and verification of asafety and security properties of multi-agent AI systems.

• Research & Development: Formalize and prove security and safety properties for agentic AI architectures and multi-agent swarms using interactive theorem provers.
• Infrastructure & Tooling: Build and scale automated proof and verification workflows designed to run on high-performance GPU clusters.
• Collaborative Research: Work closely with co-advisors, PhD/Master's students, and external collaborators (such as the Lean FRO) to integrate state-of-the-art interactive theorem proving with modern AI systems.
• Publication & Dissemination: Publish high-impact research in top-tier security, formal methods, and AI conferences/journals (e.g., CAV, PLDI, S&P, CCS, NeurIPS, ICML).

• Education:
  
  Ph.D. in Computer Science, Applied Mathematics, or a related quantitative field (completed or near completion).
• Domain Expertise:
  Strong background in formal verification, programming languages, software security, or trustworthy AI.
• Interactive Theorem Proving:
  Hands-on experience using Lean, or any similar interactive theorem proving system.
• Domain Knowledge:
  Understanding of agentic AI models, multi-agent frameworks, or AI swarms.
• Systems & Programming:
  Proficiency in functional programming (e.g., Lean) and Python/C++ for integrating verification tools with AI frameworks.

• Experience with LLM-guided formal reasoning or neuro-symbolic methods (e.g., LLM + Lean integration).
• Background in multi-agent orchestration, swarm intelligence, or adversarial robustness in security systems.
• Familiarity with distributed computing or running large-scale automated reasoning workloads on GPU clusters.

Columbia University is an Equal Opportunity Employer / Disability / Veteran

The salary of the finalist selected for this role will be set based on a variety of factors, including but not limited to departmental budgets, qualifications, experience, education, licenses, specialty, and training.

The above hiring range represents the University’s good faith and reasonable estimate of the range of possible compensation at the time of posting.