Computer Science graduate focused on full-stack software engineering, AI-assisted developer tools, healthcare technology, data-driven applications, and product-focused engineering.
I enjoy building practical software that connects clean interfaces, reliable backend systems, structured data, and thoughtful user workflows. My strongest work is across React, TypeScript, Node.js, PostgreSQL, AI tooling, mobile development, and healthcare/product platforms.
- Computer Science graduate with a concentration in Software Engineering from Arizona State University
- Interested in full-stack engineering, frontend/product engineering, AI automation, healthcare software, internal tools, and data-driven systems
- Strongest technologies: TypeScript, JavaScript, React, Node.js, Express, PostgreSQL, Python, Swift, SQL, Git, and REST APIs
- I like building products end to end, from user flows and UI to APIs, databases, validation, testing, and documentation
- Currently exploring AI-assisted development, LLM workflows, workflow automation, observability, and reliable software systems
Full-stack healthcare coordination platform
Industry-sponsored capstone platform for patients, caregivers, clinicians, and admins.
Tech: React, Vite, Node.js, Express, Prisma, PostgreSQL, Recharts
Highlights:
- Built role-based healthcare workflows across patients, caregivers, clinicians, and admins
- Implemented RBAC, OTP verification, authenticated sessions, caregiver/MPOA linking, privacy controls, and audit visibility
- Developed scheduling, messaging, medication tracking, vitals, notifications, feedback, and visit lifecycle workflows
- Built admin analytics dashboards with KPI tracking, DAU monitoring, searchable logs, filters, and pagination
- Optimized large-dataset query time and network transfer by 95 percent through pagination and backend query improvements
AI-assisted developer tooling for safer code generation
A TypeScript VS Code extension that reduces unsafe AI-generated code edits by grounding model output before workspace mutation.
Tech: TypeScript, VS Code Extension API, Zod, Vitest, LLM APIs
Highlights:
- Extracts AI agent assumptions into a typed Repo Belief Graph
- Uses a Confidence-to-Action Gate to route changes to proceed, inspect, ask, or block
- Integrates LLM plan extraction, workspace scanning, evidence grounding, Zod validation, structured patch generation, and per-file diff review
- Validates behavior with Vitest, property-based tests, and benchmark scenarios
- Built around responsible AI-assisted development, output validation, and developer trust
On-device ML recovery intelligence iOS app
Privacy-first iOS recovery app that analyzes biomechanical movement using pose estimation and deterministic scoring.
Tech: Swift, SwiftUI, MVVM, SwiftData, HealthKit, QuickPose, MediaPipe, SIMD3
Highlights:
- Computes 50+ biomechanical metrics from 33-point pose data at 120Hz
- Scores squat, hip hinge, posture, balance, range of motion, and asymmetry
- Uses NaN-safe joint-angle calculations and fallback logic for incomplete pose data
- Built a 12K+ line SwiftUI/MVVM client across 59 files
- Designed a reusable 700+ line SwiftUI design system powering 20+ views
Swipeable civic discovery and action platform
A deployed civic engagement platform that helps users discover, understand, and act on local policy issues.
Tech: Next.js, TypeScript, Prisma, PostgreSQL, Redis, Clerk, Claude
Highlights:
- Built feed, saved, profile, discussion, and issue-tracking workflows
- Implemented Redis-backed state consistency and low-latency APIs
- Built personalized ranking using interests, district matching, geography, and engagement velocity
- Integrated Claude-powered summarization for dense civic issue content
Published AI/ML research project
Machine learning research project using Spotify API data to classify songs by emotional tone.
Tech: Python, Keras, scikit-learn, Pandas, NumPy, Seaborn, Matplotlib, Spotify API
Highlights:
- Built a neural network for music emotion classification
- Used feature preprocessing, scaling, label encoding, K-Fold cross validation, and confusion-matrix reporting
- Achieved 76 percent classification accuracy
- Published findings on acousticness, valence, tempo, and emotional tone classification
- Full-stack applications with real users and clear workflows
- AI-assisted tools that improve developer productivity and trust
- Healthcare and clinical workflow software
- Internal tools, dashboards, and automation systems
- Data-driven applications with SQL-backed models
- Frontend experiences that make complex systems easier to use
- Reliable APIs, validation layers, and testable software
- AI automation and LLM-powered workflows
- Full-stack product engineering with React, TypeScript, and PostgreSQL
- Backend API design and data modeling
- Testing, validation, and software quality
- Cloud fundamentals and deployment workflows
- Healthcare, education, and developer tooling products
Taught Java programming, 3D printing, character modeling, video production, and ethical AI through hands-on student projects.
Researched music emotion classification and built a Keras/Spotify API machine learning model with Python and scikit-learn.
Analyzed Google Analytics, AdSense, YouTube, Instagram, and Facebook performance data to support growth and content optimization.
Primary Focus: Full-stack software engineering, AI tools, healthcare/product workflows
Frontend: React, Next.js, TypeScript, JavaScript, Tailwind CSS, SwiftUI
Backend: Node.js, Express, REST APIs, Prisma, PostgreSQL, MySQL, Redis
AI/ML: LLM APIs, Claude, Gemini, Keras, scikit-learn, Pandas, NumPy
Quality: Vitest, Zod, JUnit, validation, debugging, documentation
Interests: AI automation, healthtech, education technology, developer tools, internal tools