Kabir Menghrajani, MCS
Software Engineer
Java
Python
C++
C#
TypeScript
JavaScript
HTML
CSS
AWS
Generative AI
NLP
PyTorch
Tensorflow
React
Node.js
Django
SQL
PostgreSQL
Bash
Linux Command Line
Git
About Me
Welcome to my website! My name is Kabir, and I'm currently a software engineer at IBM. My interests include Full Stack Development, ML/AI, and Cloud Computing. Recently, I've been building generative AI apps using cutting-edge language models such as GPT-4o and Gemini 1.5. I love solving real world problems using technology - in particular, I enjoy working on projects that improve peoples lives in a meaningful way. I hold Masters and Bachelors degrees in computer science from the University of Virginia.
Outside of work and academics, I enjoy several sports including basketball, climbing, golf, and table tennis. In my free time I play chess, go on hikes, and explore new places. I also enjoy learning about all kinds of technological accomplishments, including and beyond those in computer science.
Want to chat? Feel free to contact me at any time.
📧: kabir.meng@gmail.com
📞: +1 (703) 687-7545
Education
Master of Computer Science
University of Virginia, August 2022 - December 2023
Relevant Coursework:
Machine Learning, Natural Language Processing, Cloud Computing, AI for Social Good, Design and Analysis of Algorithms, Network Security and Privacy, Learning for Interactive Robots, Software Security via Program Analysis, Human-Computer Interaction, Cyber-Physical Systems
Bachelor of Science, Computer Science
University of Virginia, August 2019 - December 2022
Relevant Coursework:
Algorithms, Program and Data Representation, Computer Architecture, Operating Systems, Cybersecurity, Database Systems, Computer Networks, Network Security, Cloud Computing, Machine Learning, Theory of Computation, Advanced Software Development, Digital Logic Design, Discrete Mathematics, Robotics for Software Engineers
Experience
Software Engineer
IBM | Reston, VA | January 2024 - Present
- Built a user-friendly unified visual interface in React, enabling users with minimal coding experience to build applications on low-code platforms such as Salesforce and Appian
- Designed and implemented a microservices architecture hosted on AWS, enhancing scalability and maintainability
- Spearheaded the design as the primary architect of an innovative ML-based acoustic sensor solution proposed to the USSOC
- Optimized system performance and reliability through service-level objective monitoring and improvements
Machine Learning Intern
IBM | Reston, VA | May 2023 - August 2023
- Designed a Dockerized preprocessing system to denoise and normalize gunshot audio, increasing the accuracy of the multimodal model in noisy situations by 30-300% across various features such as distance, angle, and azimuth
- Created and fine-tuned an audio denoising autoencoder with 70% accuracy in Python using TensorFlow, PyTorch, and Librosa
- Authored and presented a scientific research paper to communicate findings to 40+ stakeholders including representatives from the Department of Homeland Security and Nordic NATO
Software Developer Intern
Opexus | Gaithersburg, MD | June 2021 - August 2021
- Developed console applications in C# and an ASP.NET MVC web app to generate XML outlines of assembly files for the eCase platform, reducing creation time by 95%
- Created and presented a design document to communicate proposed design changes for a new feature to the team
Teaching Assistant, Cybersecurity
University of Virginia | Charlottesville, VA | January 2021 - May 2021
- Mentored students in foundational and intermediate cybersecurity skills during office hours twice a week and provided feedback on assignments for 40+ students
- Facilitated development of problem-solving, analysis, and hands-on experience through weekly labs and led discussions of pivotal moments in cybersecurity history through recurring Cyber Chats
Big Data Intern
Neustar | Reston, VA | July 2020 - September 2020
- Implemented Prometheus and Grafana for monitoring server metrics, reducing time to identify problematic Linux servers by 90%
- Presented project to Executive VP and CTO, winning 1st place in Intern Demo Competition
STEM Camp Instructor
Systemic Solutions | Northern VA | June 2018 - August 2018
- Taught coding, cybersecurity, rocketry, and robotics to students aged 9-17
- Guided elementary school students as they built their first EV3 robots
- Supported over 100 students developing passion and skills in STEM
Projects
Gemini Research Assistant
June 2024
- Developed a context-aware research assistant leveraging Gemini 1.5's extensive context window, allowing users to curate personalized collections of information sources (text, links, websites) and engage in focused information retrieval chats
- Built a responsive web app using TypeScript, Node.js, React, and Next.js, integrated with PostgreSQL and AWS S3
Bedtime Story Generator
May 2024
- Built a web application to generate custom bedtime stories for children, including illustrations and voiceovers using OpenAI API
- Developed on Node.js using React and the Next.js framework, integrated with a PostgreSQL database through Prisma
- Utilized Tailwindcss for styling, and integrated the Stripe API for payments and Clerk for authentication
Ensemble NLP for Mental Health Warning Sign Detection
December 2023
- Built an ensemble NLP model with NLTK and TensorFlow in Python to classify messages that may be indicative of several types of mental illnesses and identify which words in the message signal that something could be wrong, scored 98.6% accuracy across four mental illness markers
Context-Appropriate Social Interaction for Robots
December 2023
- Extended the Python model for household tasks in the Watch-And-Help paper to tasks involving social interaction in VirtualHome
Password Cracking with First-Order Markov Chains
May 2023
- Generated passwords using first-order Markov Chains in C++ to bias English words, cracking 20.7% of 11 million passwords
Verification of Autonomous Navigation Through Signal-Controlled Intersections
May 2023
- Performed verification of the safety of autonomous vehicles at signal-controlled intersections using SafeBench and CARLA
Game Theoretic Approach to a Personalized Health Recommender System
December 2022
- Prototyped a health recommendation system with Python and Java that uses principles of game theory to recommend dietary and exercise actions tailored to the specific user and their health goals, boosting healthy lifestyle adherence by ~25%
Oort: Efficient Federated Learning via Guided Participant Selection
December 2022
- Implemented Oort with FedScale to improve the performance of federated training and testing with guided participant selection
Customizable Key Bindings for Increased Keyboard Shortcut Adoption
December 2022
- Built an interface with JavaScript, HTML, and CSS to allow users to customize keyboard shortcuts and conveniently view available shortcuts, improving shortcut acceptance by ~40%
Roommate Matcher
May 2021
- Built a web application using HTML, CSS, and JavaScript on the Django framework for Python, authenticated with OAuth, integrated with Travis CI, and hosted on Heroku; matches UVA students with potential roommates based on preferences using a distance algorithm, refining match quality by ~80%
- Led a team of five, helping teammates navigate blockers while taking responsibility for the main UI and backend components
Car Mapper
September 2020
- Built a web application using HTML, CSS, JavaScript, Flask, Google Maps API, and GCP to map the transfer fees of cars listed on CarMax to a given user
Covid-19 Infection Spread Tracker
March 2020
- Built a web application using HTML, CSS, JavaScript, and MongoDB that predicts Covid-19 infection spread using TensorFlow and Keras and identifies which hospitals will be unprepared
Fast Food Nutrition Checker
August 2017
- Built a web application using HTML, CSS, and JavaScript that calculates the combined nutritional value of meals from 10 popular fast-food chains scraped with Python