A model is the visible tip; underneath sits the plumbing that decides whether it is useful or a liability. This track is about that plumbing: data engineering and warehousing to feed models, retrieval-augmented generation to ground large language models in real knowledge, and the operations and agent design that keep generative AI dependable in production. It is pitched at engineers, not newcomers.
AI and Data Engineering
The part of AI nobody demos: the infrastructure
An advanced, four-course ExpertTrack about the unglamorous half of AI: the plumbing. It covers feeding models with well-built pipelines and warehouses, grounding them in organisational knowledge through retrieval, running them reliably at scale, and orchestrating agents that act on their own.
What you will learn
- →Design data pipelines that feed LLMs through vector databases and retrieval-augmented generation
- →Judge LLM performance, safety guardrails and cost when selecting and fine-tuning models
- →Run LLMOps in production: automated deployment, monitoring and drift detection
- →Architect multi-agent systems that reason in steps and use tools autonomously
- →Join data-engineering foundations to generative AI into end-to-end systems
Curriculum
- 1Building intelligent agent architectures
2 weeks
- 2Productionising large language models
2 weeks
- 3Mastering BigQuery: building data warehouses for business performance
2 weeks
- 4AI data pipelines and knowledge systems
2 weeks
Prerequisites
- Familiarity with machine-learning or LLM concepts
- Some exposure to cloud environments
- Intermediate Python and foundational SQL
Concepts this course teaches
- Data Engineering
Pipelines and warehousing are the track's core.
- Retrieval-Augmented Generation
RAG on vector databases for enterprise apps.
- Large Language Models
Productionising and operating LLMs at scale.
- Generative AI
Advanced, infrastructure-side generative AI.
- Agentic AI
The curriculum includes agentic-AI patterns on top of its data-engineering core.
- MLOps
Covers LLMOps: the LLM-flavoured slice of MLOps practice.
FAQ
- Is this one course?
- No. It is an ExpertTrack: four linked short courses you take at your own pace, each carrying its own certificate, with a track award at the end. The courses can also be studied standalone.
- How technical is it?
- Advanced. It assumes intermediate Python, some SQL and prior exposure to ML or LLM concepts; the value sits in productionising and scaling, not in first principles.
Course facts
- Provider
- FutureLearn
- Partner
- Starweaver
- Level
- advanced
- Duration
- About 8 weeks · 1–2 hours/week
- Price
- $39 per month after a free 2-day trial
- Pace
- self-paced
- Credential
- certificate
Data engineeringData pipelinesGoogle BigQueryETLVector databasesRetrieval-augmented generationLLMOpsAgentic AI
View course at FutureLearnAffiliate disclosure: course links are affiliate links; an enrolment through one may pay Getting Digital a commission from the provider, at no extra cost to the learner, and it does not influence how courses are described.