Zohaib
ZohaibHassan16
ML engineer. Build things that learn and lie convincingly to make Turing proud.
Languages
Repos
20
Stars
0
Forks
0
Top Language
Jupyter Notebook
Loading contributions...
Repositories
20AI-powered media forensics dashbaord that uses VQA, NLI, and ELA to detect deepfakes, tampering, and logical contradictions between an image and its caption.
Open Source Semantic Layer & Knowledge Engineering Framework
No description provided.
Distributed pipeline for SEC EDGAR 10-K extraction, vectorization, and RAG-based retrieval.
AutoServe is a MLOps stack for real-time AI. It uses FastAPI and NVIDIA Triton to serve YOLO models with near-zero latency via CUDA Shared Memory.
graph based stream processing framework
Turso is an in-process SQL database, compatible with SQLite.
Helm chart to deploy Kellnr on kubernetes
A deterministic, perfect-information card game engine written in Rust. Features AI bots, Python ML bindings, and a Tauri/Svelte desktop UI.
TPU inference for vLLM, with unified JAX and PyTorch support.
OpenJudge: A Unified Framework for Holistic Evaluation and Quality Rewards
A lightweight python module for text transformation, cleaning and analysis — beginner friendly and open source.
System Level Intelligent Router for Mixture-of-Models at Cloud, Data Center and Edge
An intelligent agent that automatically curates hackathons from your email newsletters, evaluates them using AI, stores them in Google Sheets and posts to Twitter.
Standardized test-functions for optimization algorithms and machine-learning
Empty Fridge is a computer vision tool that scans your pantry photos to suggest creative, practical recipes tailored to your dietary needs
A deep learning project using Transfer Learning (MobileNetV2 vs. ResNet50) to build a fast and accurate traffic sign classifier for the GTSRB dataset.
A Reinforcement Learning study comparing Custom CNN Vs. Nature CNN architecture for VizDoom using PPO, with Grad-CAM heatmaps
Turn your whiteboard pictures into clean notes
A real-time fraud detection pipeline using an Autoencoder to monitor and flag anomalies in streaming financial transaction data.