- Full-stack web apps with React, Next.js, TypeScript, Python APIs, auth, and databases
- Interactive frontend work: clean UI, motion, 3D web experiences, and data visualization
- Applied AI systems: RAG, local inference, realtime model integration, and automation
- Physics-informed simulation and optimization for aerospace and mechanical systems
- Turbomachinery, propulsion, and computational design workflows
| Project | What it shows |
|---|---|
| Orbit Website 26 | Next.js App Router site using React 19, TypeScript, Tailwind, Prisma/Postgres, auth, Zod, and modern tooling. |
| Nano Website | React + TypeScript + Vite frontend with Three.js, React Three Fiber, Drei, Framer Motion, and polished interactive UI work. |
| DataByte WebDev Task | Browser-based AI integration task with FastAPI, WebSockets, local ONNX inference, and a live frontend prediction flow. |
| CS Minor Project | React/TypeScript pathfinding and DSA visualizer with Tailwind and interactive algorithm UI. |
| Project | What it explores |
|---|---|
| BladeLab | Physics-driven turbomachinery design environment with Flask, FastAPI, OpenEnv-style endpoints, tests, and Docker support. |
| SchwarzschildGeodesics | C++/SFML visualization of null geodesics around a Schwarzschild black hole for gravitational-lensing intuition. |
| Rag-Application | Streamlit and LangChain RAG app experimenting with local and cloud retrieval pipelines. |
| Microjet Project | Aerospace-oriented microjet and propulsion design work. |
- Frontend: React, Next.js, TypeScript, Vite, Tailwind, Framer Motion, Three.js
- Backend: FastAPI, Django, Python APIs, WebSockets, auth flows, database-backed apps
- Data and AI: PyTorch, ONNX Runtime, LangChain, RAG, local inference, model-serving flows
- Engineering: numerical methods, CAD, generative design, turbomachinery analysis
- Build habits: clean interfaces, reproducible setup, tested behavior, readable project structure
Good software should make hard ideas easier to use. Whether it is a web app, an AI workflow, or a physics simulation, I care about clear interfaces, first-principles thinking, and building things that survive contact with real users.
- GitHub: @scienstien


