Arnav Doshi
Passionate about creating innovative solutions and building amazing digital experiences
About Me
Hey there! I'm Arnav Doshi, an undergraduate student at the University of Maryland studying computer science and business. I love solving complex problems with clean, high-impact software across full‑stack, ML/AI, and cloud.
My voyage across the tech world started when I was young. As a kid, I was fascinated by computers, leading me to build my own and develop my first projects through Scratch and Weebly. Throughout these past few years, I've had the privilege of taking part in amazing technological endeavors in and out the classroom.
Some of my most recent experience includes interning at Bristol Myers Squibb (Enterprise IT, Semantic Data Products), working as a lead software engineer at Campus Coders Crew (supporting 8,000+ students), and a Data Science/ML Internship at Sensable Inc. I work with React/Next.js, TypeScript, Python, Java, Node/Express, Docker, and AWS/GCP/Azure to deliver scalable, impactful products.
My Skills
Technologies and tools I use to bring ideas to life
JavaScript
TypeScript
Python
Java
C
C++
C#
R
OCaml
React
Next.js
Tailwind CSS
HTML5
CSS3
Node.js
Express.js
MongoDB
PostgreSQL
Always learning and exploring new technologies to stay at the forefront of innovation
My Education
Academic journey and continuous learning path
B.S. in Computer Science
Minor in General Business
GPA & Honors:
GPA: 3.9/4.0 • Dean’s List (4x) • Computer Science Honors
Relevant Coursework:
Honors & Affiliations
Leadership & Involvement
- Teaching Assistant (Fall 2024)
- Data Science Research Assistant (Spring 2024)
- Campus Coders Crew
Planned M.S. in Computer Science
Graduate Studies Intent (2027 – 2028)
Focus Interests:
Actively exploring programs, preparing research portfolio, and seeking mentorship to align specialization with long‑term impact.
My academic career is far from over, and I'm excited for what lies ahead!
My Experience
Professional journey and growth in the tech industry
Software Engineer Co-Op
Calero
- Build full‑stack features across C#/.NET services (backend) and React (frontend)
- Leverage Azure cloud services; contribute within agile ceremonies
- Ship production‑grade code with code reviews and CI/CD
DigITal Accelerator Development Program Intern – Enterprise IT
Bristol Myers Squibb (Translational Medicine & Semantic Data Products Team)
- Built clinical trial analytics dashboard (20M+ entries, 30+ SEND) with Spotfire, AWS Athena, PostgreSQL, Python
- Accelerated drug development for 70+ scientists
- Partnered with stakeholders to iteratively refine KPIs, filters, and UX
- Wrote LLM prompts for database auto access provisioning; saves 5+ hrs/week of manual approvals
- Reduced consumer access wait time from hours to <5 minutes
- Shipped automated data access tool (JS SDKs + custom MCP servers, GPT‑4.1) used by 15,000+ employees for daily tasks
- Conducted unit testing and LLM prompt evaluation as part of SDLCs (AWS Lambda, CloudWatch, pytest)
- Ensured streamlined data access for 2000+ internal users
Software Engineer Lead
Campus Coders Crew
- Led full‑stack builds for 10+ orgs supporting 8,000+ students
- Coordinated agile workflows for 70+ developers across 9 teams
- Standardized PRs, branching, and release lifecycles
- Delivered MERN apps on AWS/MongoDB Atlas with resilient REST APIs
Teaching Assistant
University of Maryland Department of Computer Science
- Mentored students in office hours (concepts, debugging, project guidance)
- Evaluated exams and coursework with detailed feedback
- Proctored quizzes/exams; upheld academic integrity
Data Science / Machine Learning Intern
Sensable Inc.
- Architected CI/CD with Jenkins + Kubernetes for mobile deployments
- Designed CNN/LSTM for physiological signals (+26% accuracy vs baseline)
- Optimized MySQL schemas: −30% query time, 2× throughput
Undergraduate Data Science Researcher
Bill Fagan Lab
- Analyzed antelope heart‑rate datasets with Python/Pandas
- Detected/mitigated biologger errors → +32% data accuracy
- Built multithreaded NumPy parsers → +125% parsing speed
- Collaborated with Dr. William Fagan using Jupyter notebooks
Programming Instructor (Volunteer)
T20 Prep
- Taught introductory Java to high school students (remote)
- Guided project development and exam preparation
- Created supportive learning environment and resources
Every experience is a stepping stone towards excellence. Ready to take on new challenges and make meaningful contributions.
My Projects
A showcase of my work, creativity, and technical skills
BotCatcher – Social Media Bot Detection
Built an AI system that flags coordinated bot networks with 98.35% accuracy, leveraging graph learning to capture behavioral patterns — improving trust and safety for large social platforms.
FlightLens
Launched a flight‑intelligence platform that reveals how conflicts reshape global air traffic — from live detours and delays to density heatmaps and CO2 impact — enabling faster operational decisions for disruptions.
Personal Portfolio Website
This site: a vibrant single‑page portfolio built with Next.js 15, Tailwind CSS, Framer Motion animations, and tsparticles background. Smooth scroll navigation, responsive layout, EmailJS contact form, and deployed on Vercel.
News Article Logical Fallacy Detector AI Agent
Python-based agent using LangChain + RAG to analyze articles for logical fallacies. Pulls live news via SerperAPI, enriches context with Wikipedia scraping, and runs a two‑stage LLM chain to produce fallacy and counterfactual reports.
VoyageWatch – Sea of Thieves Event Tracker
Multiplayer web app for posting and editing in‑game events on a live Leaflet map. Real‑time updates with Socket.IO, Spring Boot backend with REST + WebSockets, MongoDB storage, Redis TTL caching, and Dockerized services.
Star Wars: Dataverse
Star Wars encyclopedia and battle simulator. SWAPI-powered searches, live character duels with Socket.IO, animated with Framer Motion, secure auth via Firebase, state with Zustand, routing via React Router, and Material UI styling.
Functional Prosthetic Hand with Real-Time Feedback
Built a functional prosthetic hand prototype with real‑time gesture recognition (Python, OpenCV). Implemented precise motor control in C++ and TensorFlow models to adapt grip strength from user feedback. Frontend control interface in React + TypeScript; deployed services on AWS EC2 with Docker; styled UI with Bootstrap and used Jupyter for data analysis/model tuning.
SonicDB – Music Industry Insights Engine
Designed a relational schema in PostgreSQL for artists/albums/genres. Integrated Kaggle + web‑scraped datasets; developed optimized SQL to extract insights on performance, trends, and sales. Used pgAdmin for administration, modeling, and query optimization to ensure consistency and performance.
