MACHINE LEARNING ENGINEER
AMINNAIM.
Building intelligent systems, data-driven solutions, and applying ML/DL to solve real-world problems.
MACHINE LEARNING · DEEP LEARNING · DATA-DRIVEN SYSTEMS
GAZA, PALESTINE31°30′N 34°28′E
- PYTORCH
- LLMS
- COMPUTER VISION
- PYTHON
- SCIKIT-LEARN
- PANDAS
- NUMPY
- OPENCV
- MEDIAPIPE
- PROMPT ENGINEERING
- FASTAPI
- NEXT.JS
- GIT
- JUPYTER
02 / SELECTED WORK
(FOUR SELECTED REPOS — ML / CV / LLM SYSTEMS)
SELECTED WORK
01 — 2026ML GenX Bioreasoning — Track A
Prompt-only LLM pipeline that predicts whether a CRISPRi knockdown shifts a target gene's expression — up, down or neutral — using biology-aware prompting and an offline knowledge base.
FIXED LLM · NO FINE-TUNINGAUROC EVALUATIONVIEW REPO
02 — 2026AI Interviewer
Full-stack SaaS platform for AI-assisted technical interviews — live voice conversation, speech-to-text with Whisper, TTS responses and candidate scoring analytics.
GPT-4.1 + WHISPER STT + TTSJWT + OAUTH ×4 PROVIDERSVIEW REPO
03 — 2026Air Draw
Draw in thin air — MediaPipe hand-tracking turns webcam gestures into neon canvas strokes you can draw, pinch-grab and erase, no mouse required.
MEDIAPIPE HANDS TRACKING4 GESTURE CONTROLSVIEW REPO
04 — 2026Medical Insurance Cost Prediction
InsuraVision AI predicts annual medical insurance costs from demographics with a scikit-learn regression model served through a FastAPI backend and Next.js frontend.
LINEAR REGRESSION MODELSCIKIT-LEARN + PANDASVIEW REPO
03 / TOOLKIT
(THE DAILY ML STACK)
TOOLKIT
- /01
Python
Core language
- /02
PyTorch / TensorFlow
Deep learning frameworks
- /03
Scikit-learn
Classical ML
- /04
Pandas / NumPy
Data wrangling
- /05
LLMs & Prompt Engineering
Applied gen-AI
- /06
Computer Vision
OpenCV · MediaPipe
- /07
Git / GitHub
Version control
- /08
Jupyter
Experimentation
04 / ABOUT ME
(THE HUMAN BEHIND THE MODELS)
ABOUT ME
31°30′N 34°28′E
GAZA CITY, PALESTINE 🇵🇸
OPEN TO — REMOTE / GLOBAL
I'm Amin — an ML engineer from Gaza building intelligent systems, data-driven solutions, and applying machine learning & deep learning to solve real-world problems.
From LLM reasoning pipelines and computer-vision interfaces to deployed prediction services, I turn messy data into products that actually work — end to end.
00
PROJECTS SHIPPED
00
TOOLS IN THE STACK
00/7
LEARNING MODE
- /01Machine Learning & Deep Learning→
- /02Data Engineering & Analysis→
- /03Computer Vision→
- /04Model Deployment→
- /05Problem-Solving at Scale→
05 / CONTACT
CONTACT
HAVE A PROJECT, ROLE OR IDEA IN MIND?