SUT NAW

PROJECTS

Image Filter

Implemented image-processing algorithms in Python, including convolution- based blurring, sharpening, Sobel edge detection, color conversion, and boundary handling.
Leveraged higher-order functions by passing filters as function objects,enabling reusable color-filter conversion and composable filter pipelines.
Developed a content-aware seam-carving algorithm using dynamic programming to calculate energy maps, identify minimum-energy seams, and resize images.

Snek Is You

engineered a rule-driven grid game in Python, implementing movement, collision detection, object transformation, and win/defeat conditions.
developed a parser that interpreted dynamic rules such as SNEK IS YOU, including compound horizontal and vertical expressions.
designed recursive algorithms to manage complex pushing and pulling interactions while maintaining valid game state.

Symbolic Algebra

Developed a symbolic algebra system using expression trees to evaluate, differentiate, simplify, and format mathematical expressions.
Applied inheritance, polymorphism, encapsulation, and operator overloading to create an extensible architecture for additional symbolic operations.

LISP Interpreter

Developed a Scheme interpreter in Python with tokenization, recursive parsing, expression evaluation, and an interactive REPL.
Implemented lexical scoping through parent-linked environment frames, variable lookup, definitions, and closures that preserve their enclosing environments.
Leveraged first-class functions and Python’s callable interface to evaluate built-in operations, construct user-defined lambda functions, and execute functions with dynamically bound arguments.

Autonomous Bot

Create bot using PID controller, Infrared line following sensor, wheel encoder, and ESP32 for dynamical systems and control

EXPERIENCES

SPARK Lab — Robotic Manipulation

Configured an AgileX dual-arm robot with two seven-joint manipulators and grippers for data collection and policy execution.
Integrated and streamlined single-arm and dual-arm inputs for training and inference with OpenPI’s π_0.5 vision-language-action policy.

Laboratory for Nuclear Science

Developed a Python pipeline to process particle-collision data, select muon events, and calculate key kinematic observables.
Implemented an ABCD method using muon isolation and impact parameters to estimate signal and background distributions through weighted histograms.