I led a security analysis of SMPTE ST 2110 networks, identifying vulnerabilities by combining risk assessment, EBU member interviews, and hands-on lab simulations. My practical testing included executing a 'Rogue Grandmaster' PTP takeover and quantifying how network attacks cause critical timing failures. This foundational research produced actionable best practices and has established the security of the NMOS API control plane as a key vulnerability, which is the focus of my ongoing work.

Certificate Authorities (CAs) are trusted to verify that an applicant for a certificate legitimately represents the domain name(s) in the certificate. Traditionally, this verification is done through various ad-hoc methods. The Automatic Certificate Management Environment (ACME) protocol (RFC8555) aims to facilitate the automation of certificate issuance by creating a standardized and machine-friendly protocol for certificate management.

The goal of this project is to implement and apply a secure multi-party computation engine (secure MPC, or SMC) in a semi-honest (passive) adversarial setting using the Python 3 programming language. We implement an SMC framework that works for generic arithmetic circuits assuming the existence of a trusted third party.

In this project, we first design and implement an anonymous authentication mechanism using attribute-based credentials.Then we conduct a privacy evaluation of a location-based service and propose and evaluate a privacy defense. Finally, we implement and evaluate a network-traffic fingerprinting attack when a user makes a location query.

Catch The Flag (CTF) is not exactly a project but rather a collection of exploits on different softwares and systems. Notable affected domains are software security, system security, binary exploitation, effective mitigations, fault isolation/privilege separation, strong sanitization, and software testing (fuzzing). For C programs, the main reverse engineering tool used was Ghidra and GDB.

Road extraction from satellite imagery is essential for applications such as urban planning and infrastructure development. In this project, we propose a machine learning- based approach to extract road networks from high-resolution satellite images, producing accurate representations of road structures. For this project task, we have a dataset of satellite images acquired from GoogleMaps. We also were provided ground-truth images where each pixel is labeled as road or background.

The project is titled "The study of correlation intractable hash functions" and was the topic of the mandatory semester project under the supervision of Prof. Serge Vaudenay and his lab LASEC. This work explores the Fiat-Shamir (FS) transform, a method for converting interactive sigma-protocols into non-interactive zero-knowledge (NIZK) proofs. We analyze the security implications of CI functions, particularly in the context of trapdoor sigma-protocols, and discuss practical constructions using shiftable shift-hiding functions. This study provides a comprehensive yet accessible foundation for researchers interested in modern cryptographic primitives and NIZK protocols.