Overview

Legacy software systems power critical infrastructure but often suffer from security vulnerabilities, outdated architectures, and high maintenance costs. Our research develops AI-driven techniques to automate the modernization of these systems, from translating unsafe C code to memory-safe Rust, to decomposing monolithic applications into microservices.

Key Directions

• C-to-Rust Translation: Neurosymbolic techniques that produce safer Rust code from C projects, preserving functionality while gaining memory safety.
• Microservice Migration: AI-guided dependency analysis and partition refinement for decomposing monolithic applications into microservice architectures.
• Translation Quality: Empirical studies measuring the quality, correctness, and idiomatic nature of machine-translated code.
• Multi-modal Specifications: Generating specifications in multiple modalities to guide and validate code translations.

Contributors

Selected Publications