Tshedup Ghissing

Tshedup Ghissing

CS Student | AI/ML, Backend & Systems | Building Real-World Software


About

Computer science student interested in AI/ML, software engineering, backend development, and systems programming. I enjoy building projects from scratch and learning by solving real technical problems. Currently exploring Python, C++, FastAPI, web development, AI tools, and computer architecture. I've worked on projects involving file transfer systems, AI-powered developer tools, and emulator development. I'm always looking to learn, build, collaborate, and turn ideas into working software.

Skills
JavaScript
TypeScript
Python
Java
React
Node.js
Google Cloud
Gemini
PostgreSQL
MongoDB
System Design
Projects

zebraAFT

zebraAFT is a lightweight, code-based file transfer application designed to move files between devices without requiring user accounts or login. Create a Drop, upload files, share the short Drop Code, and access the files from another device.

Svelte
TypeScript
Vite
Python
FastAPI
Uvicorn
psycopg2
python-dotenv
python-multipart
Supabase
PostgreSQL
Supabase Storage
Vercel
Render
GitHub
QRCode

Kinesis

MotionLens is an AI-powered, markerless motion-capture system that transforms a smartphone camera into a portable motion-capture tool. It uses local pose estimation to detect human body landmarks, processes and smooths the motion data, streams it in real time to Unity, and retargets the movement onto a 3D humanoid character. The system is designed for indie game developers, animators, students, and small studios, with potential applications in rapid animation prototyping, character animation, motion reference, and AI-assisted rotoscoping.

Python
MediaPipe Pose Landmarker
OpenCV
NumPy
WebSockets
JSON
Unity 6.2
C#
NativeWebSocket
TextMeshPro
Git
GitHub
Mixamo Humanoid Rig

Fruit Fly Neural Simulator

A browser-based 3D Drosophila neural simulation environment that uses a real 3D fruit-fly model as the physical body of the simulation. The current work focuses on building the fly's body-control system, including keyboard-controlled movement, camera-relative navigation, anatomical orientation, separate body-part identification, wing pivots, and extraction of individual leg components so they can eventually be driven by a walking controller. The long-term goal is to connect these physical body movements to a biological Drosophila connectome and create a closed-loop environment where sensory input can influence neural activity and neural activity can control the fly.

TypeScript
Vite
Three.js
GLTF/GLB
WebGL
HTML
CSS
Node.js
npm
Git
GitHub
Education

B.Sc. Computer Science with Cognitive Systems

GITAM Deemed to be University, Hyderabad

Jul 2025 to May 2028
Certifications

Claude 101

Anthropic

Jun 2026
Tshedup Ghissing · DevFest Hyderabad 2026