Jobs by Fastorial
Back to all jobs

Senior Machine Learning Engineer

0 views
AI Crew SolutionsHyderabadHybrid

5+ years

Posted today

  • AWS
  • Docker
  • Git
  • Kafka
  • Python
  • RabbitMQ

About the role

AI Crew Solutions (AICS) is hiring a Machine Learning Engineer to own the full lifecycle of our predictive models. You'll build data pipelines, develop and evaluate models, deploy them on live data, and report on how they perform. You'll also be responsible for keeping the services behind those models running in production.

Responsibilities

  • Build and improve predictive models for regression and classification, including neural networks, that run continuously on live data.
  • Operate the production services behind those models, including monitoring, incident response, and safe rollouts and rollbacks.
  • Design pipelines that combine real-time and historical data from multiple internal and external sources.
  • Evaluate models through back testing, live comparisons, drift monitoring, and metrics tied to business outcomes.
  • Keep cloud computing costs within budget.
  • Write code that is tested, documented, and easy to review.

Required qualifications

  • 5+ years of experience building and operating production ML systems, including on-call ownership.
  • Strong Python and SQL skills, with experience using pandas, NumPy, scikit-learn, and either XGBoost or LightGBM.
  • Deep experience with regression, including regularized models, tree ensembles, quantile regression, and prediction intervals.
  • Hands-on experience building neural networks in PyTorch or TensorFlow for tabular and sequence data.
  • Experience with time-series modeling and forecasting, including preventing data leakage, building point-in-time features, and designing fair evaluations.
  • Hands-on experience operating services on AWS and Linux, including troubleshooting from logs.
  • Experience with Git and CI/CD, including automated testing and deployment.
  • Clear communication skills and the ability to translate model results into decisions.

Nice to have

  • Experience with real-time or streaming data, such as Kafka, JMS, or AMQP feeds.
  • Experience with probability calibration and designing thresholds or alerts.
  • Experience working in regulated environments or with sensitive data.
  • Familiarity with MLOps tools and practices, including experiment tracking, model registries, and containers. Examples include MLflow and Docker.
  • Knowledge of airlines, aviation, or crew scheduling.