← All 186 courses
AI-217 AI & Data

Recommender Systems Engineering

Build recommenders as the two-stage systems they actually are: cheap candidate generation feeding an expensive ranker inside a fixed latency budget. Learners implement collaborative filtering and a two-tower retrieval model on the same interaction data, then confront cold start, popularity bias and the feedback loops that make a model look better offline than it performs live. The course closes with an A/B design that ties offline metrics to online lift.

B4Professional
6modules
42skill atoms
2role journeys
Curriculum

What this course covers.

6 modules, 42 named skill atoms. Expand any module to see them.

1Two-Stage Architecture7 skill atoms
candidate generation vs ranking rolesmillions to hundreds to tens funnelretrieval recall as the hard ceilinglatency budget split across stagesbusiness rules and eligibility filteringfinal re-rank for diversityindex refresh cadence
2Collaborative Filtering and Factorisation7 skill atoms
user-item interaction matriximplicit vs explicit feedbackALS and BPR objectivesSVD and matrix factorisationitem-item similarity neighbourssparsity and long-tail coverageimplicit and LightFM baselines
3Two-Tower Retrieval7 skill atoms
separate user and item encodersin-batch negative samplingsampled softmax logQ correctionshared embedding space geometryANN index built over item vectorstower staleness and re-embedding cadencefeature parity between the towers
4Features, Freshness and Cold Start7 skill atoms
batch vs real-time feature pathssession and sequence featurescontent features carrying new itemsbandit exploration for cold itemsonboarding signals for new usersfeature freshness SLAtraining serving skew from stale features
5Bias and Feedback Loops7 skill atoms
popularity bias amplificationposition bias in click logsinverse propensity weightingexposure bias baked into training datafilter bubble and diversity metricslogged-policy confoundingrandomised exploration traffic as ground truth
6Offline Metrics versus Online Lift7 skill atoms
recall at k and nDCG offlinehit rate coverage and catalogue reachwhy offline gains vanish onlineA/B design with guardrail metricsoff-policy counterfactual evaluationnovelty and serendipity measuresp99 serving latency under peak load
Where it fits

AI-217 in the role journeys.

This course appears in 2 of our 45 role journeys. Here is what a learner takes immediately before and after it in each.

AI / ML Engineer

Professional stage
AI-214AI-217AI-218

Data Scientist

Professional stage
AI-214AI-217AI-218

Roles this course serves

The capability ladder

This course is authored to band B4.

Every course we run is written to one rung of the CASI ladder, so a plan can be assembled to take a team from where they are to where they need to be.

What do B1–B6 mean?The CASI Capability Ladder — click to expand

Every course targets a band on the CASI Capability Ladder — our six-band proficiency scale, anchored to open standards (O*NET, ESCO, NICE, NIST AI RMF, Bloom's). A band tells you how deep a course goes, and what evidence proves it.

What the learner can doTypical evidence
B1
AwareUnderstands concepts and vocabulary; uses tools with guidance
Knowledge checks
B2
FoundationPerforms standard tasks correctly in familiar contexts
Guided labs, autograded exercises
B3
PractitionerDelivers complete pieces of work independently
Scenario labs, proctored hands-on exams
B4
ProfessionalHandles production-grade complexity, trade-offs and failure modes
Break-fix drills, design defenses
B5
AdvancedEngineers systems end-to-end under constraints; leads others
Rubric-scored capstones, vivas
B6
ExpertSets direction; recognised authority across teams
Portfolio + panel evaluation

A note on B6. Courses in this catalog target B1–B5. B6 is not taught — it is recognised, through a portfolio and a panel, once someone is setting direction for others. Every journey here is built to land a learner at B5.

Next step

Run AI-217 for your team.

This course runs at several lengths depending on how deep you need to go and how much of it your people already have. Tell us who is being trained and we will scope it.

Add it to a training plan Talk to our team Check your team’s level free