hoangsonww/Spot-the-Scam-AI-Job-Fraud

🎒 An AI/ML-powered, full-stack job-posting fraud copilot delivering calibrated risk scores, natural-language rationales, interactive chatbot, and real-time insights - powered by ONNX/MLflow exports, classical + transformer ensembles, quantization, Optuna hyperparameter tuning, Google AI LLM, and a sleek Next.js dashboard.

43
/ 100
Emerging

Combines XGBoost, LightGBM, and fine-tuned DistilBERT into a calibrated ensemble routed through FastAPI, with artifact versioning via MLflow to eliminate train-serve skew. Implements a gray-zone policy that flags ambiguous predictions for human review rather than forcing binary decisions, paired with a feedback loop that captures override signals for continuous retraining. Deploys across AWS, Azure, Google Cloud, and Oracle via Kubernetes and Terraform, with Argo CD for GitOps-driven rollouts and full observability through test calibration metrics (ECE=0.0066).

No Package No Dependents
Maintenance 13 / 25
Adoption 5 / 25
Maturity 9 / 25
Community 16 / 25

How are scores calculated?

Stars

11

Forks

7

Language

Python

License

MIT

Last pushed

Mar 13, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/hoangsonww/Spot-the-Scam-AI-Job-Fraud"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.