Requirements

Expected technologies

Jira

AI

Agile

Our requirements

• Education: Bachelors/Masters in CS or equivalent industry experience (5+ years in ML, 2+ in GenAI).
• Certifications: CSM, PSM, or SAFe Scrum Master.
• Technical Fluency:
• Understanding of AI/ML workflows (model training, validation, deployment).
• Familiarity with tools: Azure DevOps, Jira, MLflow, Databricks.
• Awareness of cloud AI services (Azure ML, GCP Vertex AI, AWS SageMaker).
• Soft Skills:
• Conflict resolution in high-ambiguity AI projects.
• Stakeholder management (C-suite to junior data engineers).

Your responsibilities

• Agile Leadership for AI Teams
• Lead sprint planning/retrospectives tailored to AI project cycles:
• Account for model training times, hyperparameter tuning, and data dependencies.
• Manage backlog refinement for AI-specific tasks (e.g., data labeling, A/B testing).
• Resolve AI workflow bottlenecks:
• GPU resource contention, dataset versioning delays, or model validation bottlenecks.
• Coordinate with DevOps for MLOps pipeline integration (CI/CD for models).
• Stakeholder Collaboration
• Bridge technical/non-technical communication:
• Help translate AI model metrics (precision, recall, F1) into business impact for stakeholders.
• Manage expectations around probabilistic AI outputs (e.g., hallucination risks).
• Partner with AI Architects/Data Engineers to:
• Align sprints with cloud resource availability (Azure ML, GCP Vertex AI).
• Track dependencies for multi-team initiatives (e.g., GenAI + data lake teams).
• Process Optimization
• Adapt Scrum/Kanban for AI/ML realities:
• Define “Done” for model iterations (e.g., validation accuracy ≥ 90%).
• Implement hybrid sprints for research (exploratory) vs. deployment phases.
• Introduce tools for AI Agile:
• Jira plug-ins for experiment tracking (MLflow, Weights & Biases).
• Burndown charts accounting for cloud compute costs (Azure/GCP spend).
• Team Enablement
• Coach teams on AI-aware Agile practices:
• Balancing technical debt (legacy models) vs. innovation (LLM prototyping).
• Psychological safety for failed experiments (e.g., model drift issues).
• Foster continuous learning:
• Organize workshops on AI ethics, MLOps, or GenAI trends.
• Share retrospectives across AI teams to propagate best practices.