Example of building a Flask REST API for a classifier model. The same process can be applied to other machine learning or deep learning models once you have trained and saved them.

Aim was to build neural network, machine learning models and heuristic algorithm. As well as build upon that REST API and make in form of Docker container. Main part for recruitment process for ML Engineer intern.

Deploying an ML model on Heroku involves creating a web application that serves predictions. Package the model with necessary dependencies, build a REST API using Flask or FastAPI, deploy on Heroku using Git integration, ensuring proper environment configuration and scale resources as needed.

market-intel-ml is a Python-based MLOps project for analyzing Finnish companies using open data from PRH and StatFin. It builds, evaluates, and serves a supervised machine learning model for estimating the probability of company closure within 24 months, with a rule-based Company Context Score retained as a baseline via a REST API.