I am Rutvij, a CS grad student at UT Dallas.
My specialization is in Intelligent Systems. Eventually, I would like to pursue a PhD in CS & work as a researcher.
Drop down each section to learn more about me!
Interests ...
My interests are far too wide ranging for me to ever catch up with them. Some of them are:
- Machine Learning, especially Reinforcement Learning
- Logic Programming & AI - automated reasoning!
- Data Science, particulary the statisitcal & mathematical side of it.
- Programming Language Design & Compiler Theory
- Functional Programming - Haskell & Lisp in particular
- Theory of Computation & Algorithm Design
- Numerical Optimizaiton & Computational Geometry
- System Programming - OSes, Databases etc.
- GUI Programming (native apps) - MVC, UI/UX design
And thus I call myself a deluded dilettante!
๐ญ๐ฑ Iโm currently working on and learning ...
- Implementing the first interpreter in Crafting Interpreters using Python. The implement repo for PyLox.
- Learning Lisp (Scheme) & a new way of looking at fundamentals of our craft from SICP. My notes.
- Learning Go & making practical projects following prompts at Gophercises.
- Grokking C language through the lens of its creators K&R2.
๐ค I'm looking forward to learning these in the near future ...
- Learning Quantum Algorithms and Neural Net Mathematics this Fall!
- Haskell - through a category theoric lens & revisiting LYAH before learning from new resources.
- Reinforcement Learning & Neural Networks in greater depth.
Artificial Intelligence & Machine Learning
-
Capstone project involving analysis of Transistor Load Pull Data for Qorvo. [Data Analysis]
-
Custom implementation of Decisiton Trees as a part of a graduate ML course. [ML]
-
Contact tracing data analysis [ML]
- Custom implementations of Bagging & Boosting (adaBoost) ensembles on ID3 trees.
- Custom implementation of a simple NN
- Use of various sampling methods and a comparison of their effectiveness at addressing class imbalance.
- Check out the Project Report for details!
-
Statistical analysis projects using R Language as a part of a grad course. [Statistics]
-
Lane Line Detection: Detecting lane lines on roads & stop signs using Open CV. [Computer Vision]
-
Medial Axist Transform: Medial axis transform using Delaunay triangulations. MAT was used to demonstrate too of its applications, one, to compress shapes by reducing a dimension. Another, a computer vision cross-application to perform pattern matching and find similar shapes using path matching. [Computational Geometry & Computer Vision]
-
Implementation of A* search using Python to solve 8-Puzzle game and comparison with other search methods.
-
A Q&A system using Prolog, which can answer basic questions given an encoded description of a simplified real world environment. [Logic Programming]
-
Flower Images Classification - Transfer learning on trained CNN models to classify flowers.
-
Cheque Perspective Transform - A cheque scanning workflow with amount extraction using HTR.
iOS & GUI App Development
- Flix: An app that allows users to browse movies from the The Movie Database API. [iOS]
- TwitterClone: A basic twitter app to read your tweets, compose new tweets, and like + retweet timeline tweets. [iOS]
- Parstagram: An Instagram clone with a custom Parse backend that allows a user to post photos, view a global photos feed, and add comments. [iOS]
- Used PyQT as a part of my internship & for my DB course to build a contacts list app using OBDC (CRUD, triggers, text search etc.). [PyQT]
Database Engine
- An SQLite like DB, named RioDB, implemented as a part of a Database Design course.
C/C++ Programming
- Few examples created as a part of the coursework for an OS course
- Few more as a part of an Unix Programming Course.
Cybersecurity: Docker Wordpress vs Kali
- As a part of a Cybersecurity course I took with Codepath, we created an airgapped system to learn and perform pentesting.
- The course initially suggested the use of Vagrant to setup and maintain the VMs for testing (a Kali Linux instance and a WordPress server).
- I replicated the process using Docker, making it easier & faster for first time learners to deploy the test enviroments with minimum fuss.
- Also leveraging the light weight nature of containers to make the process less resource intensive.
HDL Programming
- Built an 16-bit ALU using the Verilog HDL and an interactive simulation Logism.