Mohammed Yusuf Ahmed

Building AI systems that actually work in production.

Full-stack and AI systems engineer focused on low-latency inference, modular architecture, and real-world deployment.

AVAILABLE FOR WORK

India

Tech Stack

Frontend

3
React
Next.js
Tailwind

Backend

3
FastAPI
Node.js
SQLite

AI/ML

4
PyTorch
MediaPipe
OpenCV
Whisper

Systems

3
RAG
OCR
Multimodal

I like to make things interactive.

Recent Visitors

No entries yet.

View all →

Typing Game

The quick brown fox jumps over the lazy dog.

WPM

0

Best

0

Experience

Built Whirl, a real-time productivity system with on-device posture detection (MediaPipe + OpenCV), optimizing for low-latency inference and eliminating cloud dependency

Developed Agenza, a modular AI agent marketplace using Next.js, implementing dynamic agent orchestration, prompt pipelines, and scalable frontend architecture

Engineered Vecna 3D, a 2D-to-3D reconstruction pipeline using Python and computer vision techniques, automating mesh generation from image inputs

Designed and deployed AI-first full-stack systems, focusing on performance, modularity, and clean UI/UX with production-oriented architecture

Built and experimented with local AI pipelines (RAG, multimodal inputs, OCR, speech), emphasizing privacy-preserving and offline-capable intelligence systems

Featured Projects

Latest Writing