Data Science - Administrator

Overview

Data Science role focused on architecting and building scalable AI-powered applications, ML/GenAI solutions, and agent-based workflows that move beyond proof of concept into production.

What you'll do

• Architect and build AI-powered applications and services designed to scale beyond a proof of concept
• Fine-tune and productionize ML/GenAI models, integrating them into existing products, workflows, and APIs
• Design retrieval-augmented generation and agent-based workflows for real business use cases
• Make sound engineering trade-offs on system design, performance, and maintainability as usage grows

What you'll need

• Strong software engineering fundamentals in Python, including writing clean, modular, testable code and designing maintainable codebases/architectures, not just scripting models
• Proven experience building and deploying ML/GenAI solutions end-to-end, from model/prompt design through to production deployment, using frameworks such as scikit-learn, TensorFlow, or PyTorch
• Deep knowledge of LLMs and generative AI, including prompt engineering, retrieval-augmented generation (RAG), embeddings, and vector databases
• Experience designing and building AI agents and automation workflows (not just calling APIs; architecting multi-step, tool-using systems)
• Backend and API development experience, with the ability to integrate AI models cleanly into existing products and services
• Cloud platform experience (Azure and/or AWS) and comfort deploying containerized workloads (Docker/Kubernetes) at scale
• Solid grasp of the full ML lifecycle: data preprocessing, model evaluation, monitoring, and deployment, with an eye toward reliability and scalability in production, not just notebook experimentation
• Git and standard version control practices

Nice to have

• Familiarity with MLOps tooling and practices (CI/CD for ML, model versioning, monitoring/observability)
• Experience with data pipelines and ETL processes, and working with both structured and unstructured data
• Understanding of responsible AI, privacy, and security practices in AI systems
• Experience with MCP (Model Context Protocol) tools or similar emerging agent-tooling standards

Details

• Hybrid work: colleagues are expected to be in their local office or working onsite with clients at least three days per week, with an anchor day each week for office-based teams