I am speaking at NixCon North America this year. Reach out if you’d like to meet with me at SCaLE 21x (March 14-17, 2024 in Pasadena, California).
I am a fourth year undergraduate at UC San Diego double majoring in Math and Computer Science and minoring in Classical Studies. Outside of class, you will find me at the Triton Unmanned Aerial Systems club working working as a jack-of-all-trades and on 3D path planning for our autonomous plane. I mentor FIRST Robotics Team 812 where I hope to inspire high schoolers with a “sense of wonder” through hands on experience.
I am currently working with Professor Pat Pannuto on the Rust-based Tock Operating System. Tock takes advantage of Rust type-safety to offer fault isolation, dynamic memory management, and concurrency that are not typically available to power constrained microcontrollers. We are revisiting network interface design and abstractions. Our goal is to integrate fair sharing of limited radio on-time and network bandwidth. I am particularly interested in how interfaces can be used to make compile-time guarantees about network operation. Previously, I was a member of The Computing for Social Good Lab at UCSD where I researched mental burnout and problem gambling on social media using NLP.
I am not actively searching for new positions at this time. I am interested in work where I can work on systems software (Operating Systems, Low-Level, or Performance) and prefer to be able to open source my work. Please check my resume for links to my past experiences.
Software keeps getting larger and harder to understand. Currently, there is no easy way to check if a system is meeting its specifications or to confirm assumptions about its behavior. The only options are unreliable manual testing or to create costly formal proofs to match code’s behavior with a specification. I believe this unmanageable complexity is already stifling innovation and will be the biggest problem in future software. Thus, the most impactful and urgent area of software engineering is to make verifying large systems practical. Lowering barriers to making safe software will encourage better software. I dream of a world where software was designed to be observable, debuggable, and safe.
I am interested in investigating the applications machine learning techniques and its intersection with various fields, particularly NLP. I see computers’ knowledge of human language as the next big interface for humans to utilize computing power more naturally. I look forward to computers gaining a semantic understanding of language.
I am also passionate about the application of simulation in developing robust systems. I am currently looking into using Blender simulations to accelerate software development on aerial vehicles. Simulations can be integrated into automated testing pipelines and used for human sanity-check verification during rapid development cycles.
👋 Hey, I’m Samir, but you can call me Samir.
My recent covid hobby has been an Among Us addiction. I also have been going through my backlog of things I’ve been meaning to do (like giving this site a little love 🫀). If you have any recommendations, I enjoy discovering new music and seeing all the interesting genres people come up with. Despite my true love being software and me being wholly incompetent at the physical engineering disciplines, I enjoy fiddling around with hardware. Being a student consumes a lot of my time, but I hope to spend more time working on physical projects in the future.
I’d love to talk about anything! Feel free to suggest a time that works here 🕰.
Things I love
- ⚙ understanding complex systems
- 💻 keeping an organized workspace against relentless entropy
- 🤔 shows & movies that make you think
- 📕 sci-fi: Neuromancer is a recent recommendation
- 👀 privacy
- 🎵 discovering new music
- 🏃♂️ XC running
- 💾 people who do proper backups
- 🪄 the magic when things just work
- 🪐 intellectualizing at night
- 📰 current favorite Wikipedia article
- 🦍 current favorite game
- 🐳 current favorite movie
For more info
Message me through any of the contacts in the sidebar. Email is preferable.
I would love to hear your feedback, constructive or otherwise. As an effort for personal CI/CD and behavioral refactoring, I have made this anonymous feedback box.