Responsibilities
- Writing services and APIs that allow access to the ML models;
- Reducing the run-time of processing millions of data points in multiple ML services;
- Integrating ML services with multiple parts of our product infrastructure that is being supported by multiple teams;
- Communicating with the tech leads of other departments to ensure the best alignment on the implementation and release;
- Implementing the resilience logic for the services (metrics, alerts, retries, fallbacks, throttling, health checks, auto-recovery);
- Supporting and improving internal tools for the engineers and data annotators;
- Creating a system design for the new services.
Requirements
- Experience as an ML Engineer for 3+ years;
- Strong proficiency in Python and knowledge of common ML libraries (PyTorch, transformers, pandas, polars, catboost, etc.);
- Experience with TensorRT;
- Experience in developing and maintaining APIs and microservices;
- Solid understanding of system design and architecture principles;
- Knowledge of containerization and orchestration (Docker, k8s).
Qualification That Can Be a Plus
- Experience with ML orchestration systems (Kubeflow, ClearML, BentoML, etc.);
- Experience with GPGPU (CUDA) or any other massive parallel programming;
- Experience with resilience engineering practices;
- Experience with Triton inference server.
Conditions
- A steep springboard for personal and professional growth;
- Developing your professional competencies through courses and/or conferences;
- Language courses, mindfulness webinars;
- Modern MacBook;
- The possibility of self-realization and the ability to influence technical decision-making;
- A large friendly community, international IT teams, corporate events, team building and hackathons.