Production-Ready Salary Prediction MLOps API

An end-to-end MLOps system for census-income classification, built as a reproducible pipeline rather than a one-off notebook. It combines a cross-validated Random Forest model with DVC and AWS S3 artifact tracking, slice-based fairness evaluation, a typed FastAPI inference service, automated tests, GitHub Actions CI, and gated deployment to Heroku. The model achieved 0.685 test F1, with documented subgroup metrics and explicit responsible-use boundaries for a transparent, production-minded ML workflow.