Data science education has a gap that tutorials and competitions do not fill: the messy reality of working with unreliable data, unclear stakeholder requests, and systems that do not behave like textbook examples. The Analyst, a browser-based simulator, tries to close that gap by placing learners inside a fictional company with nine interconnected assignments and a rotating set of weekly briefs.
Meridian Living Systems and the connected estate
The simulator is set at Meridian Living Systems, a multi-region company that sells connected-home products and subscriptions, dispatches field service technicians, runs warehouses and commerce operations, and supports customers after purchase. Meridian has grown through acquisitions, platform migrations, and operational changes, and its data reflects that history. Nothing is clean, and nothing is isolated.
Across nine assignments, learners move between different analyst roles and teams within the same company. The systems, customers, devices, branches, definitions, and consequences belong to one connected estate, not nine unrelated classroom datasets. That continuity matters because real analytical work rarely starts from scratch. You inherit someone else's schema, someone else's metric definitions, and someone else's unresolved problems.
Nine assignments with escalating complexity
The sequence begins with metric reconciliation. Two satisfaction figures, 7.6 and 3.8, are headed for the same executive review, and the analyst must reconcile their scales, populations, and coverage before choosing which number to present. It is the kind of problem that does not require advanced modeling but demands careful SQL and honest judgment about what the data actually represents.
Assignment two asks the analyst to certify Q2 orders, revenue, and fulfillment timing after an acquisition cutover, where two systems were stitched together and the seams show. Assignment three evaluates whether a mobile-navigation experiment warrants a full rollout, requiring the analyst to work through experimental design and statistical significance under real constraints. Assignment four puts the analyst in a crisis scenario: a storm has disrupted operations, and the analyst must recommend global rollback, scoped containment, or monitored continuation based on incoming service performance data.
The later assignments move into production-grade model risk, where the stakes are higher and the data problems are harder to spot. Complexity rises throughout, but the simulator expects judgment to begin in assignment one.
Weekly briefs and operating conditions
Beyond the nine core assignments, The Analyst runs a permanent rotation of 16 Priority Briefs that change weekly. These briefs simulate the ongoing, time-pressured work that fills an analyst's calendar between major projects: ad hoc requests, conflicting priorities, incomplete data, and stakeholders who need answers now.
The simulator deliberately includes what clean tutorials remove: unclear requests from non-technical stakeholders, competing deadlines, unreliable data grains, operational limits on what can be queried or changed, and accountable handoffs where the analyst's work gets reviewed by someone else. These conditions are not obstacles to the learning. They are the learning.
A self-guided manual with built-in honesty
The Analyst ships with a self-guided manual that supplies readiness gates, honest workload routes, a repeatable assignment cadence, stuck protocols for when progress stalls, and a spoiler-controlled way to review your own work. The manual does not give away answers. It gives away process, which is the thing most data science education underweights.
For instructors, a separate desk provides planning guidance, workload bands, assessment boundaries, and local review tools. The separation is intentional: the learner route stays focused on the work, and the instructor route stays focused on facilitation and evaluation.
The Career Readiness Evidence Guide connects assignments to bounded NACE competency practice, portfolio review, résumé language that accurately reflects what the learner can do, interview preparation, and ready-to-use career-center formats. The goal is not just to teach SQL or statistics but to produce evidence that the learner can do the work under realistic conditions.
Why simulation matters for data science hiring
The gap between what data science education teaches and what employers need is well documented. Candidates can solve LeetCode problems and train models on clean datasets, but struggle when asked to reconcile conflicting metrics from two systems, explain a result to a non-technical stakeholder, or decide what to do when the data does not support a confident answer. The Analyst targets that gap directly, using a connected fictional company to simulate the ambiguity and accountability that define the actual job.
The simulator runs in any modern browser with no installation required. Assignments are accessible immediately, and the weekly brief rotation means the environment stays active as a practice tool beyond the initial sequence.