Eason Kamander

Created three robust data collection tools in Python and Rust, fixing major accuracy errors and delays, simplifying workflows, and drastically improving performance.

Expanded the SQL schema, providing access to disease data across historical versions, and leveraging design fundamentals to improve performance and eliminate redundancy.

Improved online map visualizations, adding features while refactoring a 5,000 line script into a few short static files.

Architected a comprehensive SQL schema for representing real-time electrical data, reducing server latency from minutes to milliseconds, and easily eliminating a wide variety of errors.

Engineered a high-performance data pipeline in Python, ensuring that the data collection process can effortlessly adapt as new electrical devices and signals are incorporated, and enabling a successful migration within 48 hours.

Created a user-friendly web portal, enabling clients to design their own security requirements.

Emphasized a modular approach, making it easy for clients to reuse and share the filters they've made, and for organizations to seamlessly provide access to new sets of metrics.

Implemented a simple compiler, safely and efficiently transforming these security constraints into the SQL queries SPHERE uses to detect anamolies.

Building on my masters research project, I have been designing a type system for the pure lambda calculus, incorporating intersection types, first-class co-induction, variance via logical duality, and using Kruskal's Tree Theorem to guarantee decidable approximations of undecidable type inference algorithms.

These ideas are very exciting to me because they provide a natural way of representing and reasoning about complex programs, helping to push the limits of proof-theoretic capabilities in formal logic.

Developed an extremely rudimentary operating system in C, with an emphasis on efficiently handling interrupts, and providing support for a variety of IO devices.

Implemented some simple test programs, including a memorization game and an interpreter for a simple functional language.

Experimented with hand-written assembly, and custom allocation strategies in Rust.

TabSpace helps users declutter their browser, making it easy to find, bookmark, and close tabs once the browser's tab bar gets too crowded. It provides an intuitive and powerful interface for selecting and managing groups of tabs all at once.

Since its first release in 2017, TabSpace has drastically evolved alongside my progression as a programmer. More recently, major browsers have improved their built-in tab management solutions, mitigating many of the problems I hoped to address. Nonetheless, this project was very pivotal for me, and still has a few hundred users.

Mentored students throughout their programming practicals, answering questions, providing feedback, and marking solutions in four of my favorite courses: Functional Programming, Principles of Programming Languages, Concurrency, and Compilers.

Nominated for a departmental award by my students, who cited my “passion for gaining a deeper understanding of the underlying concepts” and “encouragement to improve code beyond the requirements.”

Led weekly meetings after school (and online during the pandemic), teaching lessons I'd prepared on math topics from solving geometry puzzles to understanding calculus concepts.

Often inspired by math videos on YouTube, focused on building visual intuition, providing an alternative to our memorization heavy curriculum.

Led our NYSML competition team to qualify for states for the first time in years.

Worked as a math tutor (first as a volunteer after school and later independently) writing specialized workbooks and teaching lessons aimed at gradually building visual intuition.

Gained a better appreciation for what I like about math, and learned how to present beautiful ideas without blindly appealing to math as an authority figure.

Created one of my first websites to advertise my tutoring business.