Add Differential Privacy to a Mental-Health App's Analytics Dashboard
Overview
What this challenge is about.
Add Laplace/Gaussian noise to a mental-health dashboard, track privacy budgets, block unsafe queries, and compare noisy vs true answers. Earn a verifiable certificate.
The scenario
Digital-therapeutics companies ship software treatments that are reviewed and cleared by regulators like prescription products, so their engagement data is treated as protected health information. Re-identifying even one patient from an internal dashboard is both a privacy breach and a regulatory incident.
The Brief
What you'll do, and what you'll demonstrate.
Retrofit a clinical-engagement analytics dashboard with differential privacy that preserves usefulness for twelve common queries while bounding how much any single dashboard user can learn about an individual patient.
Earning criteria — what you'll demonstrate
- Choose and correctly parameterize Laplace and Gaussian mechanisms for counts, sums, and averages using the OpenDP API
- Design and implement an adaptive composition tracker that enforces a per-user privacy budget across repeated queries
- Quantify the utility cost of differential privacy and communicate it with confidence intervals
- Reason about residual re-identification risk for small clinical cohorts and decide when a query must be blocked
- Explain differential-privacy guarantees in plain English to a non-technical privacy executive
Program Fit
Where this fits in your program.
Sharpens the same skills your degree expects you to demonstrate.
Aligned coursework coming soon.
Skills
Skills you'll demonstrate.
Each one shows up on your verified credential.
- Differential Privacy
Apply differential privacy to solve real industry problems and demonstrate production-level capability.
- Privacy Budget
Apply privacy budget to solve real industry problems and demonstrate production-level capability.
- Python Programming
Apply python programming to solve real industry problems and demonstrate production-level capability.
- Data Analytics
Apply data analytics to solve real industry problems and demonstrate production-level capability.
- Privacy Engineering
Apply privacy engineering to solve real industry problems and demonstrate production-level capability.
- Threat Modeling
Apply threat modeling to solve real industry problems and demonstrate production-level capability.
Careers
Career paths this challenge builds toward
Completing this challenge demonstrates skills that transfer directly to these roles:
Privacy Engineer
This challenge mirrors the core privacy-engineering job: turning a legal re-identification concern into a quantified, enforced technical control. You leave able to design privacy budgets, implement noise mechanisms, and defend the trade-offs to leadership.
This challenge sharpens
- differential-privacy
- privacy-budget
- privacy-engineering
Security Engineer, Data Protection
Retrofitting a live analytics system with bounded privacy loss and a small-cohort gate is exactly the defense-in-depth work data-protection security engineers own. You practice threat-modeling re-identification and shipping the guardrails in code.
This challenge sharpens
- threat-modeling
- privacy-engineering
- python-programming
Analytics Engineer, Regulated Data
Balancing dashboard usefulness against privacy is a daily tension in regulated analytics. By measuring noise-induced error and proving twelve queries stay trustworthy, you build the analytics judgment regulated-data teams hire for.
This challenge sharpens
- data-analytics
- differential-privacy
- python-programming