Purpose & Overall Relevance for the Organization:A Machine Learning Engineer is a specialist who applies its expertise in end-to-end Model Development Lifecycle (MDLC), artificial intelligence and engineering - software, DevOps, data, cloud, and platform; to improve and productionize state of the art machine learning models. They can pivot between data centric (data engineering) or model centric (data science) approaches to enhance predictive model performance and apply various software engineering techniques to deploy and scale models.
Key Responsibilities:Machine Learning Engineering:- Builds components for Data platform for distributed data processing pipelines and scalable feature stores including data health monitoring and alerts.
- Builds components for ML platform to enable distributed model training and evaluations, including model observability and model performance monitoring
- Designs end to end Machine Learning Pipeline (i.e MLOps)
- Works with data scientists and data engineers to productionize data pipelines and machine learning models, so that various business requirements can be implemented and scaled
- Assists in generating last mile data readiness (for example embeddings) for the Data Scientists so that they can quick move towards value generation i.e. by directly applying models on curated features
Analytics:- Applies a range of machine-learning techniques in consultation with data scientists and domain experts to enhance models within the explainable AI(XAI), performance and responsible AI constraints
- Selects, acquires, and integrates features (AI focused data components) for analysis.
- Applies unsupervised ML techniques (like clustering) to data for unknown pattern identification and to run precursory analysis for supervised ML tasks.
Data management, modelling and design:- Applies exploratory data analysis (EDA), data design, data modelling and quality assurance techniques to establish, modify or maintain highly curated features for the task of AI engineering
- Fills in for all the data engineering needs or assists data engineering towards the goal of project delivery
- Implements physical database & data warehouse designs to support feature availability for MDLC
- Assists in providing accessibility, retrievability, security and protection of data in an ethical manner.
Programming/software development:- Designs, codes, verifies, tests, documents, amends, and refactors moderately complex programs/scripts.
- Develops and deploys feature engineering and model training/inferencing code with CI/CD practices in mind
- Builds cloud/on-prem native MDLC templates to orchestrate and channelize development processes for various engineering teams engaged in MDLC
- Cross applies model development/deployment in distributed processing and big data paradigms to cover for data volume, velocity, and variety constraints
Data visualization and storytelling:- Applies a of variety visualization techniques and designs the content and appearance of data visuals for storytelling and EDA
- Operationalizes and automates activities for efficient and timely production of data visuals via operationalized dashboards and reports.
- Communicates results of unsupervised learning techniques (like clustering) to identify and communicate unknown patterns in data
Testing:- Reviews requirements and specifications and defines test conditions.
- Designs test cases and test scripts under own direction, mapping back to pre-determined criteria, recording and reporting outcomes.
- Analyses and reports test activities and results.
- Identifies and reports issues and risks associated with own work.
- Embeds unit, integration and regression test cases within CI/CD processes driving MDLC
If required: People Management / Resource Management:- May be involved and gives some input on hiring Transition decisions
- Ensures appropriate leadership skills are present at every level through creating a motivational and supportive work environment in which employees are coached, trained and provided with career opportunities through development
- Allocates the different work to the respective employees considering experience, complexity, workload and organizational efficiency
- Continuously monitors and evaluates team workload and organizational efficiency with the support of IT systems, data and analysis and team feedback and makes appropriate changes to meet business needs.
- Provides team members/direct reports with clear direction and targets that are aligned with business needs and GIT objectives.
Requisite Education and Experience / Minimum Qualifications:- Bachelors Degree in Computer Science, Mathematics or similar field; Master’s degree is a plus
- 4 years’ hands-on experience as a Machine Learning Engineer or similar role, experience with financial or demand planning data is a plus
- Understanding of data structures, data modeling and software architecture
- Experience with production level MLOps - feature engineering, distributed model training, serving & inference, etc.
- Deep knowledge algorithms and Big Data technologies: Apache Kafka, Apache Spark, AWS EMR; Databricks is a plus
- Passionate and ability to write robust code in Python, R, Java
- Familiarity with machine learning frameworks (like Keras or PyTorch) and ML libraries (like scikit-learn)
- Experience with machine learning algorithms, tools (e.g., MLflow, AWS Sagemaker, TensorFlow), deep learning and/or natural language processing.
- Impeccable and to-the-point written and oral communication skills (English)
- Comfortable in presenting complex topics to stakeholders
- Proven team player who can collaborate across functions and organizations
- High resilience and solution-oriented attitude
adidas celebrates diversity, supports inclusiveness and encourages individual expression in our workplace. We do not tolerate the harassment or discrimination toward any of our applicants or employees. We are an equal opportunity employer.