I am a Software Engineer and recent graduate of UC Berkeley's EECS program. I specialize in high-concurrency backend infrastructure and AI data pipelines. I have many interests including Computer Science, film photography, graphic design, circuit design, marine conservation and much more. Throughout my life I have always and still am working towards my passions.
In distributed systems, securing service-to-service communication is critical. Agent Authenticator is a production-grade gateway built in Go that implements secure JWT (RS256) identity verification and concurrency-safe traffic control. It features a custom token-bucket rate limiter to prevent abuse and layered middlewares for real-time anomaly detection, ensuring that only authorized and behaving agents can access protected resources.
Secure data transmission across untrusted networks requires rigorous cryptographic primitives. This project is a high-performance file sharing system built in Go, leveraging AES-GCM for symmetric encryption, RSA for secure key exchange, and Argon2 for robust password hashing. It ensures total data confidentiality and integrity during transit, actively mitigating man-in-the-middle attacks through strict, low-level cryptographic protocols and secure memory management.
Autonomous aviation safety relies on precise, real-time spatial awareness in highly complex environments. During my internship, I developed a LiDAR-based object segmentation pipeline using CUDA-accelerated models to identify commercial aircraft, ground vehicles, and personnel. By extracting raw point-cloud data from ROS bags and adapting pre-trained weights from autonomous driving datasets (like Waymo), I successfully bridged the domain gap to achieve high-fidelity recognition on custom airport runway data.
