Software Engineering • Machine Learning • AI
Arham V. Doshi
University of Texas at Austin
B.S. Computer Science + Linguistics
Austin, Texas
Expected May 2028
Software Engineering • Machine Learning • AI
University of Texas at Austin
B.S. Computer Science + Linguistics
Austin, Texas
Expected May 2028
B.S. Computer Science + Linguistics
Austin, Texas
GPA: N/A
High School Diploma
Southlake, Texas
AP Computer Science A/Principles, AP Calculus AB/BC, AP Statistics
Transcribing and analyzing 4000 verbal responses from epilepsy patients performing arithmetic tasks to study cognitive-linguistic processing
Annotated and extracted acoustic features in Praat, enabling precise prosody analysis for epilepsy patient study
Transcribed morphemes and syntax using LLMs, to analyze the neuro system — improved model by 12%
Bridging the Socioeconomic Gap in Education: A Hybrid AI and Human Annotation Approach
Published a paper to CoNLL-2025 investigating socioeconomic status prediction using Reddit data, focusing on linguistic patterns in low-SES posts
Designed a custom ranking system and engineered NLP pipelines using Logistic Regression, Naive Bayes, and Transformer-based models to classify posts with improved interpretability
Preprocessed and analyzed 30,000+ Reddit entries using Python (NLTK, SpaCy), applying bigrams, trigrams, keyword expansion, sentiment analysis, and LDA topic modeling to uncover latent psychological themes
Designed a college match platform that used machine learning to deliver personalized results, integrating speech-to-text and a Siri-style UI for accessibility
Delivered a top-3 final product pitch to mentors and peers, showcasing the platform's features, user flow, and future roadmap through a live demo and slide presentation
Collaborating with Dr. Kuan Chuen Wu on Project BALTO, developing advanced machine learning models for predicting animal health and enhancing veterinary diagnostics
Applied NLP techniques to analyze medical literature and health records, enabling data-driven strategies in animal health management
Conducted data cleaning, preprocessing, and feature engineering to improve accuracy of predictive models for animal health diagnostics
Contributing to AI-powered diagnostic tools providing accurate health insights for veterinary applications
Started as primary coder programming the robot in C++, progressed to design specialist role responsible for designing pneumatics, catapult mechanism, and overall robot structure
Managed and maintained engineering notebook documenting all design processes, strategies, and coding implementations, contributing to winning design awards
Consistently ranked in top 10 in every tournament throughout 11th grade and qualified for VEX Robotics World Championship
Founded nonprofit integrating AI-powered web/app tools for multilingual education; scaled to 500+ monthly users across 4 languages.
Authored and published "The Loss in AI Translation: Examining Cultural and Linguistic Gaps in Hindi–English Neural Models" on ResearchGate.
Designed college match platform using machine learning to deliver personalized results, integrating speech-to-text and Siri-style UI for accessibility.
University of Michigan
Coursera
ID: CG2GKQGNMS45
DeepLearning.AI, Stanford University
Coursera
ID: MGGHPG92FZYF
Karlsruhe Institute of Technology
Coursera
ID: GYEGFS59CN63
I'm always interested in discussing AI, NLP research, and opportunities for collaboration.