MLOps / ML Architect – 01 starts August
- 14 – 16 years of total experience, 3 – 5 years of relevant experience on Sagemaker AI
- Hands on experience of the following –
- Structure a ML system (batch, real-time, or LLM) as modular ML pipelines that can be independently developed, tested, and operated
- Ensure the consistency of feature data between offline training and online operations
- Govern data in a feature store and promote collaboration between teams with a feature store
- Implement MLOps principles of automated testing, versioning, and monitoring of features and models.
- The modeling skills required for ML:
- How to train ML models from (time-series) tabular data in a feature store
- How to personalize LLMs using fine-tuning and RAG
- How to validate models using evaluation data from a feature store
- Identify and develop reusable model-independent features
- Identify and develop model-dependent features
- Identify and develop on-demand (real-time) features
- validate feature data, test feature functions and test ML pipelines
- Schedule feature pipelines and batch inference pipelines
- Deploy real-time models, connected to a feature store
- Log and monitor features and models with a feature store
- Develop user-interfaces to ML systems