About the role
As a Machine Learning Engineer at TheWebVale, you will construct predictive models, recommendation algorithms, and anomaly detection pipelines. You will analyze business data for retail brands, coaching networks, and property companies to forecast demand, classify records, and uncover operational trends.
You will collaborate with backend developers and data engineers within our regular two-to-three week release cycles. You will package models into lightweight API endpoints, write data cleaning scripts, and ensure dependable inference performance in production.
What you will do
- Train, evaluate, and fine-tune machine learning models on structured business datasets.
- Build automated data preprocessing and feature engineering pipelines in Python.
- Deploy trained models as low-latency microservice endpoints using FastAPI or Docker.
- Evaluate model performance using cross-validation, precision, recall, and ROC curves.
- Monitor production inference drift and schedule model retraining intervals.
- Collaborate with product designers to present predictive insights clearly to end users.
What you will need
- Hold 2–4 years of practical experience developing and deploying machine learning models.
- Demonstrate strong proficiency in Python, NumPy, Pandas, Scikit-learn, and SQL.
- Show practical familiarity with deep learning frameworks like PyTorch or TensorFlow.
- Understand foundational statistics, probability distributions, and regression techniques.
- Possess experience building RESTful model serving APIs with FastAPI or Flask.
- Communicate statistical findings and model certainty levels clearly to project leads.
Nice to have
- Experience with XGBoost, LightGBM, or CatBoost on tabular business datasets.
- Familiarity with time-series forecasting libraries such as Prophet or Statsmodels.
- Knowledge of containerizing machine learning pipelines using Docker and Linux.