Arcadia 21 час назад
Senior Applied ML Engineer
- Удалённо
- Опыт от 4 лет
- Полная занятость
PythonSQLPostgreSQLGitC#MLPyTorchpandas.NET CoreVue.jsNumPyTensorFlowScikit-learnWeb APINeural NetworksEF CoreSPAPiniaVue3React RouterVuexComposition API
Описание
Responsibilities
- develop and improve ML models for predicting the probability of purchase at different prices
- optimize pricing decisions
- build neural network models
- design controlled studies within the permissible price range, including randomization rules, probabilities of assigning price options, and safety constraints
- design A/B tests to compare new pricing policies with current ones, define primary and guardrail metrics, stopping criteria, and rollout conditions
- manage the full ML lifecycle: training, validation, inference, periodic retraining, model versioning, and feedback loop between predictions and actual results
- implement an end-to-end ML prototype with minimal HTTP API, data storage, and simulation of an external booking system
- correctly log assigned prices, experimental variants, purchase facts, and business metrics
Requirements
- solid knowledge of Python, NumPy, pandas, scikit-learn, PyTorch or TensorFlow
- proficiency in C# for backend integration
- experience in developing, training, and applying ML models, including neural networks, feature engineering, and working with tabular data
- understanding of the mathematical foundations of ML, probability theory, statistics, optimization methods, and A/B testing principles
- ability to design controlled exploration, including randomization rules and primary and guardrail metrics for pricing experiments
- experience in organizing the full ML cycle (training, validation, inference, retraining, versioning, feedback)
- strong SQL, PostgreSQL, Git, and Docker skills
- experience in developing Python applications with autotests
Nice to have
- experience in dynamic pricing and revenue optimization (demand forecasting, price elasticity, willingness to pay)
- familiarity with causal inference and sequential decision-making methods (contextual bandits, off-policy evaluation, IPW, doubly robust estimation)
- experience with MLOps tools (MLflow, orchestration, monitoring)
- experience in operating ML services
- experience in aviation, e-commerce, or revenue management
We offer
- remote work
- corporate laptop
- official employment under the Labor Code of the Russian Federation with a white salary
- Voluntary Medical Insurance (DMS)
- 100% sick pay
- opportunities for professional growth